Use of receptor chimeras to identify small molecules with high affinity for the dynorphin A binding domain of the κ opioid receptor

Article abstract of DOI:10.1016/j.bmcl.2007.11.116

A series of 2-substituted sulfamoyl arylacetamides of general structure 2 were prepared as potent κ opioid receptor agonists and the affinities of these compounds for opioid and chimeric receptors were compared with those of dynorphin A. Compounds 2e and 2i were identified as non-peptide small molecules that bound to chimeras 3 and 4 with high affinities similar to dynorphin A, resulting in K_i values of 1.5 and 1.2 nM and 1.3 and 2.2 nM, respectively.

Full text of DOI:10.1016/j.bmcl.2007.11.116

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 3667–3671
Use of receptor chimeras to identify small molecules with high
affinity for the dynorphin A binding domain of the j opioid receptor
Virendra Kumar,* Deqi Guo, Michael Marella, Joel A. Cassel, Robert N. DeHaven,
Jeffrey D. Daubert and Erik Mansson
Adolor Corporation, 700 Pennsylvania Drive, Exton, PA 19341, USA
Received 12 September 2007; revised 23 October 2007; accepted 28 November 2007
Available online 4 December 2007
Abstract—A series of 2-substituted sulfamoyl arylacetamides of general structure 2 were prepared as potent j opioid receptor ago-
nists and the affinities of these compounds for opioid and chimeric receptors were compared with those of dynorphin A. Compounds
2e and 2i were identified as non-peptide small molecules that bound to chimeras 3 and 4 with high affinities similar to dynorphin A,
resulting in Ki values of 1.5 and 1.2 nM and 1.3 and 2.2 nM, respectively.
Ó 2007 Published by Elsevier Ltd.
The identification of the mu (l), delta (d), and kappa (j)
subtypes of the opioid receptor led to the suggestions
that agonists selective for receptor subtypes might be
effective analgesics with fewer serious side effects.1 Even
though the arylacetamide series of j opioid receptor
agonists lack l opioid receptor-mediated side effects,
the utility of these agonists as antinociceptive agents is
limited due to side effects such as dysphoria, diuresis,
and psychotomimesis.2–4 In clinical trials, the naturally
occurring peptide j opioid receptor agonist, dynorphin
A, mediates analgesia without dysphoria, diuresis, and
psychosis, indicating that the antinociceptive effects of
j opioid receptor agonists could be dissociated from
their side effects.5,6 A metabolically stable analog of
dynorphin A, E2078, is an effective analgesic in post-sur-
gical patients at doses that produce no side effects.7 This
exemplifies that there are opportunities for identifying
metabolically stable small peptides or small molecule j
opioid receptor agonists as effective analgesics that lack
the side effect profile of the arylacetamides.
receptors. These different modes of binding have led to a
hypothesis that different domain selectivity of agonists
that bind to the j receptor might be related to different
patterns of side effects.9 Therefore, in an effort to dis-
cover small molecule j opioid receptor agonists that
have a therapeutic profile similar to that of dynorphin
A and related compounds, we have recently described10
the design and construction of two l/j chimeric recep-
tors composed primarily of amino acid residues derived
from the l opioid receptor for the screening of
compounds.
The chimeric receptors used in this study include one of
the chimeric receptors used in the earlier study (desig-
nated chimera 3)10 and another chimeric receptor (desig-
nated chimera 4). These receptors are depicted in Figure
1 in which filled circles represent amino acids derived
from the j opioid receptor and open circles represent
amino acids derived from the l opioid receptor. For chi-
mera 3, the 25 amino acids of the putative second extra-
cellular loop of the l opioid receptor were replaced with
the 28 amino acids (8 identical) of the putative second
extracellular loop of the j opioid receptor. The chimera
4 construct was made using a synthetic oligonucleotide
corresponding to the Bcl1–Sty1 region (343 bp) of the
human j opioid receptor in which amino acid numbers
86–178 were replaced with the corresponding amino
acids of the human l opioid receptor. This construct is
a human j opioid receptor where the first and second
intracellular loops, the first extracellular loop, and the
second and third transmembrane regions were replaced
This distinction in the side effect profiles of arylaceta-
mides and dynorphin A could in part be related to the
different binding regions for the j opioid receptor8,9
which were observed through the use of chimeric recep-
tors composed of sequences derived from j and l opioid
Keywords:
receptors.
j Opioid receptor agonists; Dynorphin A; Chimeric
*
Corresponding author. Tel.: +1 484 595 1060; fax: +1 484 595
1573; e-mail: vkumar@adolor.com
0960-894X/$ - see front matter Ó 2007 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2007.11.116

3668
V. Kumar et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3667–3671
Figure 1. Schematic representation of chimeric receptors.
with the corresponding regions from the human l opi-
oid receptor. DNA sequencing was used to verify each
construct.
Our initial approach was to introduce 4,5-dimethoxy or
4,5-methylenedioxy groups in compound 1, and vary
only the amine portion of the 2-sulfamoyl group and
the substitution in the 3 position of the pyrrolidine.
We have synthesized a novel series of 2-substituted sul-
famoyl arylacetamides of general structure 2 as potent j
opioid receptor agonists and compared the affinities of
these compounds for l, d, j opioid and chimeric recep-
tors with those of dynorphin A. Once a compound from
this series having a profile similar to that of dynorphin A
is identified, in vivo testing in various analgesic models
will be performed, not only to evaluate the analgesic
properties, but, more importantly, to assess the side ef-
fect profiles.
Competitors of [3H]diprenorphine binding from a vari-
ety of structural classes bound to these chimeras with
affinities similar to those with which they bound to the
l opioid receptor. In contrast, dynorphin A analogs
bound to the chimeras with the affinities close to those
with which they bound to the j opioid receptor. Phar-
macological characterization of [35S]GTPcS binding
mediated by chimera 3 showed that it behaved as if it
were a l opioid receptor with high affinity for dynorphin
A analogs.10 These two chimeric receptors were used to
screen for compounds that bind to the j opioid receptor
in a dynorphin-like fashion. The compounds will be
used to test the hypothesis that binding domain selectiv-
ity can be used as a guide in identifying j opioid recep-
tor selective agonists as analgesics with reduced side
effect profiles.
The diamines 3 and 3a (Scheme 1) were prepared
according to published methods from (S)-phenylgly-
cine.11,12 As mentioned earlier, the 4,5-dimethoxy
and 4,5-methylenedioxy substituted phenylacetic acids
were selected to take advantage of the electron rich
Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Ile-Arg-Pro-Lys-Leu-Lys-Trp-Asp-Asn-Gln
1
8
17
Dynorphin A
R2
R1
5
4
O
6
O
X
1
N
N
3
N
N
2
SO2NHCH3
SO2R
CH3
CH3
1
2
(X = H, OH)
After screening of non-peptide j opioid receptor ago-
nists of different templates (Upjohn, Glaxo, ICI, and
Dupont), compound 1 of the ICI template, having a 2-
sulfamoyl substitution in the phenylacetamide moiety,
was identified as a lead because it bound to chimera 3
and chimera 4 receptors with Ki values of 400 nM and
phenyl ring in providing the regioselective syntheses
of the arylacetamide portion of the target
compounds.
A general synthetic pathway was designed for the con-
densation of diamines 3 and 3a with the sulfamoyl phen-
ylacetic acid 8 in the presence of EDCI/HOBT/Hunig’s
base (Scheme 1). The desired compounds 2 were purified
by chromatographic methods and converted to either
hydrochloric or methanesulfonic acid salt for the final
isolation. The yields of these compounds are shown in
Table 1.
110 nM, respectively, while having
a
Ki value
>1000 nM at the l opioid receptor. Another observation
was made during the evaluation of these compounds
that phenylacetic acids with electron rich groups such
as methoxy or dimethoxy tended to have higher affinities
for the chimeras than unsubstituted compounds.

V. Kumar et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3667–3671
3669
R1
CO2H
R1
R2
1.
R2
8
O
SO2R
X
X
N
N
N
HN
CH3
2. EDCI/HOBT/
[(CH3)2CH]2NC2H5
. HCl or CH3SO3H
SO2R
CH3
3. HCl or CH3SO3H
2 (X = H, OH)
3; X = H
3a
R = amines (Table1)
; X = OH
Scheme 1. General synthesis of targets (example).
Table 1. Opioid and chimeric receptor binding
R2
R1
5
4
6
O
X
1
N
3
N
2
SO2R
CH3
2 (X = H, OH)
Compound R1R2
R
X
Yield (%) j Ki (nM) l Ki (nM) d Ki (nM) j-Mediated Chimera
Chimera
[
35S]GTPcS 3 Ki (nM) 4 Ki (nM)
EC50 (nM)
ICI 199441 (1)
Dynorphin A (2)
NHCH3
0.044
0.21
8.4
53
9.4
24
5.9
0.30
1.1
14
1.0
0.31
400
300
3.2
6.4
3.3
0.18
110
120
2.5
1
H
H
H
H
>1000
>1000
42
>1000
500
56
11
6.0
2a
2b
2c
2d
OCH3
OCH3
OCH3
OCH3
H at 2-Position
N(CH3)2
N(CH3)2
65
55
2.8
0.50
1.6
0.26
0.24
0.54
OH 45
46
25
21
9.3
1.5
2.8
4.8
N(CH3)CH2Ph
H
0.68
2e
2f
2g
2h
2i
OCH3
OCH3
OCH3
OCH3
OCH3
OCH3
H
67
0.31
1.1
14
5.9
52
27
15
9.0
5.4
0.85
23
0.11
0.12
0.39
0.71
0.17
0.34
1.5
38
1.2
8.3
6.8
ND
2.2
ND
N
N
OH 49
62
OH 74
75
OH 77
H
1.3
4.8
ND
1.3
ND
N
N
N
N
NCH3
NCH3
0
3.8
4.4
9.4
1.8
H
0.55
1.4
2j
0
2k
2l
O–CH2–O H at 2-Position
O–CH2–O N(CH3)2
O–CH2–O N(CH3)2
H
H
70
48
0.70
0.22
0.54
0.19
450
47
580
110
6.2
3.4
0.43
0.10
0.11
0.12
220
5.2
4.5
5.5
68
0.71
0.50
1.7
2m
2n
OH 41
82
34
6.5
O–CH2–O N(CH3)CH2Ph
H
Notes: A series of concentrations of each compound was tested for its ability to inhibit [3H]-diprenorphine binding to j, l, d, and chimeric receptors
and Ki values determined as described in Refs. 14,15 The values shown are geometric means of at least 3 determinations. The ability of compounds to
stimulate j opioid receptor-mediated [35S]GTPcS was determined as described in Ref. 16. ND = not done.
The sulfamoyl acids (8) were prepared by a slight mod-
ification of published method.13 The preparation of
these acids is described in Scheme 2. The regioselectivity
of sufamoyl substituents was directed by the electronic
and steric properties of the methylphenylacetates (4).
The mixtures of the ester 7 were readily separated by
column chromatography and each was characterized be-
fore hydrolysis to give 2-sulfamoyl acids 8.
Our attempts to prepare the 4,5-dimethoxy substituted
analog of compound 1 failed due to the presence of a
reactive proton on the sulfamoyl nitrogen. Although
the 2-sulfamoyl phenylacetic acids (9) were readily syn-
thesized, the condensation reactions with diamine 3 or
3a failed to give the desired products 10. In the presence
of condensing agents such as EDCI, these acids under-
went intramolecular cyclization to give the benzothiaz-

3670
V. Kumar et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3667–3671
R1
R1
R1
CO2CH3
CO2CH3
CO2CH3
1. Ac2O/H2SO4
SOCl2/DMF
65-81%
R2
R2
R2
2. KOAc/Ethanol
SO2Cl
SO3K
4
60-90%
R1
6
5
R1
CO2CH3
CO2H
1. LiOH/THF:CH3OH(1:1)
R2
Amines
75-88%
R2
2. H3+O
80-95%
SO2R
7
SO2R
8
R = amines (Table1)
Scheme 2. General method for preparing sulfamoyl aryl acetic acids.
H3CO
CO2H
H3CO
OCH3
H3CO
O
SO2NHR
X
X
9
N
N
HN
CH3
N
EDCI/HOBT/
[(CH3)2CH]2NC2H5
CH3
SO2NHR
10
3; X = H
3a; X = OH
H3CO
H3CO
O
NR
~85%
S
O2
11
Scheme 3. Condensation of sulfamoyl phenylacetic acids (9).
ones (11, R = CH3) in excellent yields (>85%) and the
diamines were recovered quantitatively (Scheme 3). To
prevent the formation of benzothiazones, during these
reactions, the sulfamoyl phenylacetic derivatives in Ta-
ble 1 were prepared only from secondary amines lacking
the reactive proton.
affinities.17,18 However, the sulfamoyl series of com-
pounds showed reduced j opioid receptor binding affin-
ities and gains in binding affinities for the l and d opioid
receptors. These changes in binding affinity were more
pronounced for the d opioid receptor than for the other
opioid receptors (Table 1).
The new compounds were initially evaluated in vitro for
opioid receptor binding affinities. Determinations of
affinities for j, l, and d opioid receptors were conducted
by displacement of bound [3H]diprenorphine from mem-
branes prepared from cells expressing the cloned human
opioid receptors using previously described meth-
ods.14,15 Opioid receptor-mediated stimulation of
[35S]GTPcS binding16 was used to determine functional
activities of the lead compounds (Table 1). Compounds
that showed at least 10-fold higher affinity for the j opi-
oid receptor than for the l opioid receptor were evalu-
ated for chimeric receptor affinities. The data are
presented in Table 1.
The 4,5-dimethoxy- or 4,5-methylenedioxyphenylacetic
acid derivatives (2a and 2k), which lack the 2-sulfamoyl
substitution, bind to the j opioid receptor with Ki values
of 2.8 nM and 0.70 nM and with good selectivity over
the l and d opioid receptors (Table 1). However, 2a
and 2k exhibited lower affinities for chimera 3 (Ki, 300
and 220 nM) and chimera 4 (Ki, 120 and 68 nM, Table
1). The introduction of a 2-dimethyl aminosulfamoyl
group into 2a and 2k gave compounds 2b and 2l. The
j opioid receptor binding affinities were increased 6-
and 3-fold, respectively, and there were substantial gains
in the affinities for chimeras 3 and 4 yielding Ki values of
3.2 and 2.5 nM, respectively, for compound 2b (Table
1). The selectivity with regard to the l and d opioid
receptors was, however, reduced. Even though the chi-
meric receptor affinities of compounds 2b and 2l were
approximately 10- to 4- fold lower than those of dynor-
phin A, these were the first non-peptide compounds of
the series which bound to the chimeric receptor with
The Ki values for binding to the j opioid receptor for the
new compounds ranged from 0.19 to 3.8 nM with vary-
ing affinities for l and d opioid receptors. Compounds
containing 30-S-hydroxypyrrolidine generally have been
reported to show improved j opioid receptor binding

V. Kumar et al. / Bioorg. Med. Chem. Lett. 18 (2008) 3667–3671
3671
higher affinities than would be predicted by their affini-
ties at the l opioid receptor. This may indicate interac-
tions in the chimeric binding with residues derived from
the j opioid receptor.
optimization for chimeric receptor affinities will be
undertaken.
References and notes
Further improvements in binding and selectivity were
expected upon replacement of the pyrrolidine with 30-
S-hydroxypyrrolidine, which resulted in compounds 2c
and 2m. But these and other compounds of this series
did not improve the receptor affinities and in most in-
stances the affinities were decreased relative to the corre-
sponding pyrrolidine compounds (Table 1). This
structural modification affected primarily binding to
the d opioid receptor. Thus, only a limited number of
analogs having 30-S-hydroxypyrrolidine were prepared.
1. Martin, W. R.; Eades, C. G.; Thompson, J. A.; Huppler,
R. E.; Gilbert, P. E. J. Pharmacol. Exp. Ther. 1976, 197,
517.
2. Pfeiffer, A.; Brantl, V.; Herz, A.; Emrich, H. M. Science
1986, 233, 774.
3. Dionne, R. A.; Dobbins, K. R.; Hargreaves, K. M. Clin.
Pharmacol. Ther. 1991, 49, 183.
4. Pande, A. C.; Pyke, R. E.; Greiner, M.; Wideman, G. L.;
Benjamin, R.; Pierce, M. W. Clin. Neuropharmacol. 1996,
19, 451.
5. Wen, H. L.; Mehal, Z. D.; Ong, B. H.; Ho, W. K. K.
Peptides 1987, 8, 191.
Replacement of one of the methyl groups of 2b with a
bulky group such as benzyl (2d) had no significant effect
on the binding affinities to either l or j opioid receptors.
Again, the primary effect was on the reduction in affinity
for the d opioid receptor. Similarly, no changes in the
chimeric receptor affinities were observed, indicating
available space in the binding pocket for further struc-
tural modifications at this position. The cyclic amine
derivatives of these sulfamoyl compounds, such as 2e,
bound to the opioid receptors with affinities that were
comparable to dynorphin A bound (Table 1). The affin-
ities for the chimeric receptors 3 and 4, however, were
increased, yielding Ki values of 1.5 nM and 1.2 nM,
respectively, compared to acyclic amines. Similarly the
corresponding morpholine analog 2i bound to chimeric
receptors 3 and 4 with Ki values of 1.3 and 2.2 nM. As
noted above, the corresponding 30-S-hydroxypyrrolidine
derivatives of these cyclic amines demonstrated less
desirable receptor binding profiles. The decrease in bind-
ing affinities for chimeric receptors was observed for the
N-methyl-piperazine analog (2g, Table 1) and the corre-
sponding 30-S-hydroxypyrrolidine derivatives (2h, Table
1), indicating the sensitivity of the receptors to those of
the basic functionality in this part of the molecule. The
chimeric receptor affinities of compounds having the
4,5-methylenedioxy group (2k–2n, Table 1) were compa-
rable to the dimethoxy analogs.
6. Wen, H. L.; Mehal, Z. D.; Ong, B. H.; Ho, W. K. K.;
Wen, D. Y. K. Life Sci. 1985, 37, 1213.
7. Fujimoto, K.; Momose, T. Jap. J. Anesth. 1995, 44, 1233.
8. Wang, J. B.; Johnson, P. S.; Wu, J. M.; Wang, W. F.; Uhl,
G. R. J. Biol. Chem. 1994, 269, 25966.
9. Kenakin, T. Pharmacol. Rev. 1996, 48, 413.
10. DeHaven, R. N.; Mansson, E.; Daubert, J. D.; Cassel, J.
A. Curr. Top. Med. Chem. 2005, 5, 303.
11. Costello, G. F.; James, R.; Shaw, J. S.; Slater, A. M.;
Stutchbury, N. C. J. J. Med. Chem. 1991, 34, 181.
12. Gottschlich, R.; Ackermann, K. A.; Barber, A.; Bart-
oszyk, G. D.; Greiner, H. E. Bioorg. Med. Chem. Lett.
1994, 4, 677.
13. Fraga, C. A. M.; Barreiro, E. J. J. Heterocycl. Chem. 1992,
29, 1667.
14. 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.
15. DeHaven, R. N.; DeHaven-Hudkins, D. L. In Current
Topics in Medicinal Chemistry; Enna, S. J., Williams, M.,
Ferkany, J. W., Kenakin, T., Porsolt, R. D., Sullivan, J.
P., Eds.; John Wiley and Sons: New York, 1998; pp 1.4.1–
1.4.12.
16. Schlechtingen, G.; DeHaven, R. N.; Daubert, J. D.;
Cassel, J. A.; Chung, N. N.; Schiller, P. W.; Taulane, J. P.;
Goodman, M. J. Med. Chem. 2003, 46, 2104.
17. (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.
In conclusion, compounds 2e and 2i have been identified
as small molecules having in vitro profile comparable to
that of dynorphin A. Future studies will be to evaluate
these compounds in antinociceptive models and further
18. Giardina, G.; Clarke, G. D.; Grugni, M.; Sbacchi, M.;
Vecchietti, V. Il Farmaco 1995, 50, 405.