Potent and highly selective kappa opioid receptor agonists incorporating chroman- and 2,3-dihydrobenzofuran-based constraints

Article abstract of DOI:10.1016/j.bmcl.2005.08.094

Two novel chemical classes of kappa opioid receptor agonists, chroman-2-carboxamide derivatives and 2,3-dihydrobenzofuran-2-carboxamide derivatives, were synthesized. These agents exhibited high and selective affinity for the kappa opioid receptor.

Full text of DOI:10.1016/j.bmcl.2005.08.094

Bioorganic & Medicinal Chemistry Letters 15 (2005) 5114–5119
Potent and highly selective kappa opioid receptor
agonists incorporating chroman- and
2,3-dihydrobenzofuran-based constraints
Guo-Hua Chu,a,* Minghua Gu,a Joel A. Cassel,b Serge Belanger,b Thomas M. Graczyk,b
Robert N. DeHaven,b Nathalie Conway-James,b Mike Koblish,b Patrick J. Little,b
Diane L. DeHaven-Hudkinsb and Roland E. Dollea
aDepartment of Chemistry, Adolor Corporation, 700 Pennsylvania Drive, Exton, PA 19341, USA
bDepartment of Pharmacology, Adolor Corporation, 700 Pennsylvania Drive, Exton, PA 19341, USA
Received 28 July 2005; revised 24 August 2005; accepted 24 August 2005
Available online 3 October 2005
Abstract—Two novel chemical classes of kappa opioid receptor agonists, chroman-2-carboxamide derivatives and 2,3-
dihydrobenzofuran-2-carboxamide derivatives, were synthesized. These agents exhibited high and selective affinity for the kappa
opioid receptor.
Ó 2005 Elsevier Ltd. All rights reserved.
Opioid analgesics mediate their effects through three
opioid receptor types, l, j, and d.1 Most of the opioid
analgesics at present, for example, morphine, act by
binding to the l-opioid receptor, and their analgesic
potency is associated with a spectrum of undesirable side
effects, such as physical dependence, respiratory depres-
sion, and constipation. In recent years, considerable
attention has been focused on the development of recep-
tor selective j-agonists as potent and efficacious analge-
sics devoid of the undesirable side effects of the l
analgesics.2 The most important selective j-agonists
developed so far are the arylacetamide derivatives. Since
the discovery of the one of the first selective arylaceta-
mide j-agonists, U-50,488 in the early 1970s, which dis-
played analgesic effects in vivo and did not produce
the side effects associated with the CNS, peripherally
acting j-agonists are at present the focus of interest.2,5,6
Previously, we reported the design and synthesis of a
novel series of azepinones as constrained analogs of
ICI 199441, exemplified by compound 1, which showed
high and selective j binding affinity.6 During a screening
of a combinatorial library, aryloxyacetamides (e.g., 2
and 3) were identified as highly potent j-agonists.7 Here,
we describe the synthesis and biological evaluation of
two novel series of constrained aryloxyacetamides with
the general structure I: chroman-2-carboxamides
(n = 1) and 2,3-dihydrobenzofuran-2-carboxamides
(n = 0) with various substitution groups on the phenyl
ring, especially polar substituents including acetylamino,
sulfonylamino, and carbamoyl groups. The inclusion of
such polar substituents was by design to yield com-
pounds, which may limit CNS penetration and display
peripheral selectivity in vivo. This is expected to be ben-
eficial to decrease or avoid the CNS side effects of the
centrally active kappa opioid receptor ligands while
retaining antihyperalgesic activity.2 The in vitro receptor
binding assay showed that many of these new com-
pounds possess high and selective j binding affinity.
Two representative compounds from these two novel
series of constrained aryloxyacetamides were evaluated
in the in vivo nociceptive assays and demonstrated
potent analgesic effects.
respiratory depression, constipation, or tolerance,3
a
number of related, but chemically diverse, arylacetamide
j-agonists have been reported. Among these is ICI
199441, which is 146-fold more potent than U-50,488
in vitro.4 However, these centrally acting j-agonists pro-
duced their own set of CNS side effects such as dyspho-
ria and diuresis, which prevented their further
development as analgesic therapeutics.2,5,6 To avoid
Keywords: Kappa opioid receptor agonists; Analgesic.
*
Corresponding author. Tel.: +1 484 595 1083; fax: +1 484 595
1551; e-mail: ghchu@adolor.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.08.094

G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5114–5119
5115
O
Cl
O
OH
O
OH
Cl
O
N
O
N
N
N
N
Cl
N
constrained
constrained
constrained
Cl
27
Ki (κ) = 2 nM
µ/κ: > 2500; δ/κ: 370
37
ICI 1994414
Ki (κ) = 0.043 nM
Ki (κ) = 0.125 nM
µ/κ: 480; δ/κ: 40
µ/κ: 1233; δ/κ: 558
this work
ref. 6
Cl
Cl
O
O
OH
O
N
N
N
N
X
n
I
1
n = 0: 2,3-Dihydrobenzofuran class
Ki (κ) = 0.34 nM
µ/κ: 3068; δ/κ: 1079
n = 1: Chromane class
1
The synthesis of the chroman-2-carboxamide and 2,3-
dihydrobenzofuran-2-carboxamide constrained analogs
11–26 is summarized in Scheme 1. The synthesis began
with the preparation of various substituted racemic
and enantiomerically pure (R)- and (S)-chroman-2-car-
boxylic acids and 2,3-dihydro-benzofuran-2-carboxylic
acids, compounds 6–10. Racemic chroman-2-carboxylic
acid (6: n = 0) was prepared based on a literature proce-
dure using 20-hydroxyacetophenone (4) as starting mate-
rial.8 Racemic 2,3-dihydro-benzofuran-2-carboxylic acid
(6: n = 1) was easily prepared by hydrogenation of the
commercially available benzofuran-2-carboxylic acid 5
at 60–70 psi. The corresponding enantiomerically pure
acids were obtained via chiral separation of the racemic
acids.9 The absolute stereochemistry was assigned based
on the comparison of the optical rotation data with
those reported in the literature.10,11
aromatic protons in H NMR spectra.12 Both regio-
isomers were subjected to the same derivatization and
hydrolysis sequence as described above for the synthesis
of substituted chroman-2-carboxylic acids 7 (n = 1),
to furnish the corresponding substituted 2,3-dihydro-
benzofuran-2-carboxylic acids 7 (n = 0) and 8.
The acids 9 with reversed sulfonylamino substituents at
the 6- (n = 1) or 5-position (n = 0) were synthesized via a
four-step reaction sequence from chroman-2-carboxylic
acid and 2,3-dihydro-benzofuran-2-carboxylic acid:
methyl ester formation, chlorosulfonylation with sulfur
trioxide–DMF complex/oxalyl chloride, or chlorosulf-
onic acid, reaction of the resulting sulfonyl chloride with
ammonia or methylamine, and then saponification. The
synthesis of the acids 10, or methylene analogs of the
acids 7, required six reaction steps. Iodination of
chroman-2-carboxylic acid and 2,3-dihydro-benzofu-
ran-2-carboxylic acid with benzyltrimethylammonium
dichloroiodate gave the iodides in which iodine regio-
specifically substituted at the para position to the ring
oxygen.8 Esterification of the iodoacids, followed by
treatment with copper(I) cyanide, yielded the benzonitri-
le derivatives. These intermediates were in turn subject-
ed to hydrogenation (nitrile to benzylamine) and
capping with methyl chloroformate, methanesulfonyl
chloride, and propanesulfonyl chloride to afford carba-
mates and sulfonamides, respectively, which were
hydrolyzed with lithium hydroxide to yield the acids 10.
Derivatization of the chroman-2-carboxylic acid and
2,3-dihydrobenzofuran-2-carboxylic acid was carried
out on either racemic or chiral material. Treatment of
chroman-2-carboxylic acid with nitric acid gave the 6-ni-
tro-chroman-2-carboxylic acid as a single isomer, which
was converted to the ester under standard conditions.
Reduction of the nitro group by hydrogenation followed
by reaction of the resulting aniline with acetyl chloride,
methyl chloroformate, methanesulfonyl chloride, and
propanesulfonyl chloride afforded amide, carbamate,
and sulfonamides, respectively, which were hydrolyzed
with lithium hydroxide to yield the acids 7 (n = 1). In
contrast to chroman-2-carboxylic acid, nitration of
2,3-dihydro-benzofuran-2-carboxylic acid under the
same reaction conditions yielded two regioisomers:
5-nitro-2,3-dihydrobenzofuran-2-carboxylic acid as the
major isomer and 7-nitro-2,3-dihydrobenzofuran-2-car-
boxylic acid as the minor isomer (4:1 ratio). Without
further purification, the crude mixture of acids was
esterified and the two esters readily separated by silica
gel chromatography. The structural assignment of these
two isomers was based on the coupling pattern of the
With the requisite acid set 6–10 on hand, the synthesis of
the target compounds 11–26 was conducted by coupling
the acids with 1-(2-methylamino-(S)-2-phenyl-ethyl)-
pyrrolidin-(S)-3-ol13 using either Mukaiyama reagent14
or O-(benzotriazol-1-yl)-N,N,N0,N0-tetramethyluronium
tetrafluoroborate (TBTU) as coupling reagent. In both
series, the target compounds with the (R)-configuration
at the 2-position had consistently HPLC retention times
shorter than those of the target compounds with the (S)-
configuration.

5116
G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5114–5119
A
Preparation of requisite carboxylic acids:
O
O
O
S
O
HN
OH
O
O
n
HN
R
OH
7
n = 0, 1; R = CH3CO, CH3OCO,
CH3SO2, CH3CH2CH2SO2
8
iii, iv, v, vi, vii
O
O
O
OH
O
i, ii
ref. 8, ii
OH
n
O
OH
4
6
5
n = 0, 1; Racemic and chiral acids
x, iv, xi, v, vi, vii
iv, viii, ix, vii
O
O
O
O
OH
OH
H
O
N
n
n
S
R
R1
N
O
10
9
R2
n = 0, 1; R1 = R2 = H;
R1 = H, R2 = CH3
n = 0, 1; R = CH3OCO, CH3SO2,
CH3CH2CH2SO2
Preparation of target compounds:
B
O
O
OH
O
N
xii
O
N
OH
X
X
CH3
n
n
11-26
6-10
Scheme 1. Synthesis of chroman-2-carboxylic acids and 2,3-dihydrobenzofurn-2-carboxylic acids and coupling to yield j opioid receptor agonists.
Reagents: (i) H2, 60–70 psi, 10% Pd/C, EtOAc; (ii) chiral separation; (iii) HNO3; (iv) MeOH, HCl; (v) H2, 10% Pd/C; (vi) CH3COCl or EtOCOCl or
RSO2Cl, Et3N; (vii) LiOH, MeOH-THF-H2O; (viii) HCON(CH3)2 SO3, (COCl)2 or ClSO3H; (ix) R1R2NH; (x) ZnCl2, PhCH2N(CH3)3ICl2, HOAc;
(xi) CuCN; (xii) (S)-1-((S)-2-(methylamino)-2-phenylethyl)pyrrolidin-3-ol, TBTU or Mukaiyama reagent, DCM.
The in vitro opioid receptor binding results of chrom-
anes 11–19 and 2,3-dihydrobenzofurans 20–26 are sum-
marized in Tables 1 and 2.15 In the chromane series
(Table 1), compound 11-R, the unsubstituted con-
strained analog of phenoxyacetamide 2, exhibited com-
parable high j binding affinity (Ki = 1.6 nM) but with
much greater overall selectivity versus the l and d recep-
tors (>3000-fold). In contrast, diastereomer 11-S was ca.
20-fold less active at j relative to 11-R, indicating that
the configuration at the 2-position (chromane ring) is
important for optimal j binding. This stereochemical
preference held true for all compounds in the series,
for example, 17-R > 17-S, 18-R > 18-S, and 19-R > 19-
S. The substituents at the 6-position of the phenyl ring
affected j binding. Among the different substitution
groups, the sulfonylamino groups (compounds 13, 17-
R, and 18-R) are most preferred for j affinity. Com-
pound 19-R with a carbamoylmethylene substituent also
displayed good j binding affinity and selectivity. In the
case of the reversed sulfonamide analogs, compound
15 displayed comparable affinity to 13, but increasing
the bulk on the sulfonamide nitrogen atom as in 16
led to decreased j receptor affinity versus 17-R.
In the 2,3-dihydrobenzofuran series (Table 2), com-
pounds 20-R and 20-S, the unsubstituted constrained
analogs of phenoxyacetamide 2, also possessed high and
selective j affinity. This activity is similar to that of com-
pound 11-R in the chromane series. However, in contrast
to the chromane series, the configuration at the 2-position
did not significantly influence j affinity, and in most cases,
the j binding affinity of the two diastereomeric pairs was
within a factor of 2.16 Analogous to the chromane series,
sulfonylamino groups are the most preferred substituents
for j affinity, as all the compounds having the sulfonyla-
mino moiety in the molecules including compounds 23-R,
-S, 24-R, -S, and 26-R, -S had low nanomolar j binding
affinity. The carbamoyl group (22-R, -S) was also well tol-
erated by the j receptor. The regioisomers with metha-
nesulfonylamino substituent at the 5- and 7-position,
23-R versus 24-R and 23-S versus 24-S, had comparable
j binding affinity. Similar to that observed in the chro-
mane series, the reversed sulfonamide 25 was 20-fold less
potent than sulfonamide 23-S.
In both the chromane series and 2,3-dihydrobenzofuran
series, polar substituents were generally well tolerated by

G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5114–5119
5117
Table 1. SAR of the chromane series
OH
R
N
N
CH3
Compounds
R
j Ki (nM)
l/j
d/j
CYP2D6 IC50 (nM)
O
O
O
O
11-R
1.6
>3125a
>3125b
150
1600
2900
6200
11-S
12c
30.5
17
>164a
>294a
45
>294b
224
O
O
O
O
O
N
H
13c
1.7
>2940a
O O
S
N
H
O
O
14c
31
>161a
>161b
7400
1800
O
O
N
H
O
O
O
O
15c
3.3
364
44
H2N
S
O O
16c
18.8
1.6
>266a
300
20
75
71
320
6700
9500
H
N
S
O O
O
O
O
O
O O
17-R
17-S
S
N
H
>352a
O O
S
14.2
N
H
O
O
18-R
18-S
19-R
19-S
4.5
99
>1111a
>51a
132
5
3900
>10 lM
4200
H
N
S
O
O O
O
H
N
S
O
O
O O
O
O
2.6
>1923a
>39a
615
9
H
N
O
O
130
14,000
H
N
O
O
a l Ki is estimated to be >5 lM.
b d Ki is estimated to be >5 lM.
c Mixture of diastereomers (1:1).
the j receptor. Such substituents also confer low
CYP2D6 inhibitory activity with IC50 values in the
micromolar range (Tables 1 and 2). In contrast, the kap-
pa agonists of the arylacetamide class have been shown
to bind tightly to CYP2D6, for example, ICI 199441 has
an IC50 = 26 nM against this cytochrome P450 en-

5118
G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5114–5119
Table 2. SAR of the 2,3-dihydrobenzofuran series
OH
R
N
N
CH3
Compounds
R
j Ki (nM)
l/j
d/j
CYP2D6 IC50 (nM)
250
7
O
O
20-R
0.8
>6250a
861
2
5
O
O
20-S
2.4
>2083a
>95a
250
760
770
O
O
O
21c
52.4
>95b
N
H
O
O
O
22-R
22-S
23-R
14.7
5.8
238
155
744
>340b
145
4200
4800
5400
O
O
N
H
O
O
O
N
H
O
O
O
O
O O
S
1.3
25
N
H
O O
23-S
24-R
1.2
3.9
375
167
525
14
4500
1100
S
N
H
O O
S
HN
O
O
O O
S
HN
24-S
1.3
639
26
1200
O
O
O
O
H
25c
24.5
3.6
>204a
27
93
11
N
S
O O
O
O
O
H
N
26-R
>1389a
5700
S
O O
O
H
26-S
3.3
515
29
5000
N
S
O O
a l Ki is estimated to be >5 lM.
b d Ki is estimated to be >5 lM.
c Mixture of diastereomers (1:1).
zyme.17 Cytochrome P450 enzymes serve an important
detoxification role in the body.18 Inhibition of the en-
zymes, in particular CYP2D6, may interfere with the
bodyÕs ability to metabolize xenobiotics. From a drug
development perspective, it is desirable to identify
agents that are not potent CYP2D6 inhibitors.

G.-H. Chu et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5114–5119
7. Unpublished data.
5119
Two representative compounds from these two novel ser-
ies of constrained aryloxyacetamides, chroman-2-carbox-
amide 19-R and 2,3-dihydrobenzofuran-2-carboxamide
23-S, the two diastereomerically pure compounds both
showing high and selective j binding affinity (Ki = 2.6
and 1.2 nM) and low CYP2D6 inhibitory activity
(IC50 = 4.2 and 4.5 lM)were evaluatedfor activityin vivo
nociceptive assays. Compounds 19-R and 23-S displayed
potent analgesic effects, producing 95% and 88% antino-
ciception at 300 lg given intrapaw (sc injection in dorsal
surface of paw), respectively, in the late phase formalin-
induced flinching assay.19 Compounds 19-R and 23-S also
inhibited acetic acid-induced writhing19 when adminis-
tered sc and p.o. with sc ED50 values of 0.53 and
0.75 mg/kg, respectively, and p.o. ED50 values of 1 and
3.9 mg/kg, respectively. The peripheral selectivity of these
compounds will be investigated at a latter time.
8. Ladouceur, G. H.; Connell, R. D.; Baryza, J.; Campbell,
A.-M.; Lease, T. G.; Cook, J. H. WO 99/32475, 1999.
9. The chiral separations were conducted by Chiral Tech-
nology, Inc., Exton, PA.
10. Chiral chromane-2-carboxylic acid: Schutt, H. DE
4430089, 1996.
11. Chiral 2,3-dihydrobenzofuran-2-carboxylic acid: Urban,
F. J.; Moore, B. J. Heterocycl. Chem. 1992, 29, 431.
12. 5-Nitro-2,3-dihydro-benzofuran-2-carboxylic acid methyl
1
ester: H NMR (400 MHz, CDCl3) d: 8.12 (dd, 1H), 8.08
(d, 1H), 6.93 (d, 1H), 5.35 (dd, 1H), 3.82 (s, 3H), 3.62 (dd,
1H), 3.45 (dd, 1H); 7-Nitro-2,3-dihydro-benzofuran-2-
carboxylic acid methyl ester: 1H NMR (400 MHz, CDCl3)
d: 7.92 (d, 1H), 7.43 (d, 1H), 6.98 (t, 1H), 5.45 (dd, 1H),
3.80 (s, 3H), 3.65 (dd, 1H), 3.44 (dd, 1H).
13. Synthesis and use as building block for kappa opioid
receptor agonists of 1-(2-methylamino-(S)-2-phenyl-eth-
yl)-pyrrolidin-(S)-3-ol: Zhang, W. Y.; Maycock, A. L.;
Kumar, V.; Gaul, F.; Chang, A.-C.; Guo, D.
WO00014065, 2000.
14. Mukaiyama coupling reagent mediated amide formation:
Bald, E.; Saigo, K.; Mukaiyama, T. Chem. Lett. 1975, 1,
1163.
15. (a) DeHaven, R. N.; DeHaven-Hudkins, D. L. In Current
Protocols in Pharmacology; Enna, S. J., Williams, M.,
Ferkany, J. W., Kenakin, T., Porsolt, R. D., Sullivan, J.
P., Eds.; John Wiley & Sons: New York, 1998, p 1.4.1; (b)
Raynor, K.; Kong, H.; Chen, Y.; Yasuda, K.; Yu, L.; Bell,
G. I.; Reisine, T. Pharmacological Characterization of the
Cloned j-, d-, and l-Opioid Receptors. Mol. Pharmacol.
1994, 45, 330; (c) All the compounds listed in Tables 1 and
2 were full agonists as determined in a [35S]GTPcS
functional binding assay.
In summary, two novel chemical classes of kappa opioid
receptor agonists, chroman- and 2,3-dihydrobenzofu-
ran-based constrained analogs of the aryloxyacetamides,
were synthesized and found to be potent j receptor li-
gands. Among these, eight compounds had single digit
nanomolar j binding affinity and >100-fold selectivity
over l and d. Chroman-2-carboxamide 19-R and 2,3-
dihydrobenzofuran-2-carboxamide 23-S both demon-
strated analgesic effects in the in vivo formalin-induced
nociception and acetic acid-induced writhing assays.
References and notes
16. Benzofuran-2-carboxamide 27 displays moderate j bind-
ing affinity indicating that saturation at the 2-position in
this series is important for opioid receptor binding affinity.
1. Dhawan, B. N.; Cesselin, F.; Raghubir, R.; Reisine, T.;
Bradely, P. B.; Portoghese, P. S.; Hamon, M. Pharmacol.
Rev. 1996, 48, 567.
2. DeHaven-Hudkins, D. L.; Dolle, R. E. Curr. Pharm. Des.
2004, 10, 743.
3. (a) Szmuszkovicz, J.; Von Voigtlander, P. F. J. Med.
Chem. 1982, 25, 1225; (b) Szmuszkovicz, J. Prog. Drug
Res. 1999, 52, 168; (c) Szmuszkovicz, J. Prog. Drug Res.
1999, 53, 3.
O
O
OH
N
N
27: Ki (κ) = 490 nM;
Ki (µ) > 5 µM; Ki (δ) > 5 µM
4. Costello, G. F.; James, R.; Shaw, J. S.; Slater, A. M.;
Stutchbury, N. C. J. J. Med. Chem. 1991, 34, 181.
5. Kumar, V.; Guo, D.; Cassel, J. A.; Daubert, J. D.;
DeHaven, R. N.; DeHaven-Hudkins, D. L.; Gauntner, E.
K.; Gottshall, S. L.; Greiner, S. L.; Koblish, M.; Little, P.
J.; Mansson, E.; Maycock, A. L. Bioorg. Med. Chem. Lett.
2005, 15, 1091.
6. Tuthill, P. A.; Seida, P. R.; Barker, W.; Cassel, J. A.;
Belanger, S.; DeHaven, R. N.; Koblish, M.; Gottshall, S.
L.; Little, P. J.; DeHaven-Hudkins, D. L.; Dolle, R. E.
Bioorg. Med. Chem. Lett. 2004, 14, 5693.
17. Le Bourdonnec, B.; Ajello, C. W.; Seida, P. R.; Susnow,
R. G.; Cassel, J. A.; Belanger, S.; Stabley, G. J.; DeHaven,
R. N.; DeHaven-Hudkins, D. L.; Dolle, R. E. Bioorg.
Med. Chem. Lett. 2005, 15, 2647.
18. General review of cytochrome P450 enzymes: Nelson, D.
R.; Koymans, L.; Kamataki, T.; Stegeman, J. J.; Feyer-
eisen, R.; Waxman, D. J.; Waterman, M. R.; Gotoh, O.;
Coon, M. J.; Estabrook, R. W.; Gunsalus, I. C.; Nebert,
D. W. Pharmacogenetics 1996, 6, 1.
19. For a description of the in vivo assays, see Ref. 3.