Synthesis and preliminary in vitro investigation of bivalent ligands containing homo- and heterodimeric pharmacophores at μ, δ, and κ opioid receptors

Article abstract of DOI:10.1021/jm050577x

A series of homo- and heterodimeric ligands containing κ agonist and μ agonist/antagonist pharmacophores joined by a linker chain of varying lengths was synthesized and evaluated in vitro by their binding affinity at μ, δ, and κ opioid receptors. The func

Full text of DOI:10.1021/jm050577x

256
J. Med. Chem. 2006, 49, 256-262
Synthesis and Preliminary In vitro Investigation of Bivalent Ligands Containing Homo- and
Heterodimeric Pharmacophores at µ, δ, and K Opioid Receptors
Xuemei Peng,† Brian I. Knapp,‡ Jean M. Bidlack,‡ and John L. Neumeyer*,†
Alcohol and Drug Abuse Research Center, McLean Hospital, HarVard Medical School, 115 Mill Street, Belmont, Massachusetts 02478, and
Department of Pharmacology and Physiology, School of Medicine and Dentistry, UniVersity of Rochester, Rochester, New York 14642
ReceiVed June 17, 2005
A series of homo- and heterodimeric ligands containing κ agonist and µ agonist/antagonist pharmacophores
joined by a linker chain of varying lengths was synthesized and evaluated in vitro by their binding affinity
at µ, δ, and κ opioid receptors. The functional activities of these compounds were measured in the [35S]-
GTPγS binding assay. The data suggest that the stereochemistry of the pharmacophores, the N-substituents
of the pharmacophore, ester linkages, and the spacer length were crucial factors for optimum interactions
of such ligands at opioid receptor binding sites. These novel ligands as well as their pharmacological properties
will serve as the basis for our continuing investigation of such bivalent ligands as probes of the opioid
receptor oligomerization phenomena and for in vivo studies as analgesics.
Introduction
a series of compounds containing varying length spacers linked
by either ester or amide moieties. The monovalent ligands 8
and 12 (Scheme 2 and Table 1) were also synthesized and their
µ, δ and κ affinities established for comparison with their
bivalent counterparts.
Previous reports from our laboratories indicated that mor-
phinans connected by linking spacers of varying lengths
containing identical morphinan pharmacophores such as cyclo-
rphan (1c) or butorphan (1b) were sensitive to the site of
attachment on the morphinan and the length of the linking
chain.1 It was observed that the affinity of these ligands was
sensitive to the character and length of the spacer. We have
previously reported that a compound linked by a 4-carbon chain
ester (17), a conformationally constrained fumaryl ester (19),
and a compound with a 10-carbon chain ester (18) were the
most potent ligands in this series.1 The fumaryl ester 19 was
further characterized and was shown to be a κ opioid receptor
agonist and a µ opioid receptor agonist/antagonist when
evaluated in vivo.2 The structures and affinities are shown in
Table 1. Ligands having two pharmacophores connected by a
spacer can be termed bivalent ligands and have the potential
for binding vicinal receptors.3,4 Such bridging should be
manifested by a substantial increase in potency due to the high
concentration of the pharmacophore in the vicinity of the
recognition site when the ligand is bound in a monovalent mode.
Portoghese et al.5 has reported a range of homo- and
heterodimeric ligands with varying linker lengths designed to
investigate pharmacodynamic and organizational features of
opioid receptors. For example, recently reported heterodimeric
ligands containing δ antagonist (naltrindole) and κ1 agonist (ICI-
199,441) pharmacophores joined by variable length oligoglycyl-
based linkers were demonstrated to possess significantly greater
potency and selectivity compared to their monomer congeners,
providing further evidence for the opioid receptor hetero-
oligomerization phenomena.6 It is of interest to note that most
of the cleavable ligands reported in the literature contain ester
linkers, which are designed to be cleaved by plasma esterase to
release individual pharmacophores which then can presumably
act independently on the specific opioid receptor.
Chemistry
The dextrorotatory morphinans (2b,c) were prepared from
commercially available (+)-3-hydroxy-N-methylmorphinan tar-
trate (dextrorphan) 2a, which was converted to the free base
and N-demethylated7-9 to yield the normorphinan, followed by
alkylation with cyclobutylmethyl (or cyclopropylmethyl) bro-
mide and characterization as their dextrorotatory crystalline
mandelate salts (2b and 2c), respectively. The (+),(+)-ho-
modimeric ligand 3 was prepared from 2b with sebacoyl
chloride in the presence of triethylamine. The homodimeric
ligands 4, 5, and 6 containing only the levorotatory pharma-
cophores 1d, 1e and 1b were similarly prepared. The homodi-
ester 6 was prepared by condensing 1b with muconyl chloride,
which was made from muconic acid with thionyl chloride
(Scheme 1).10a-c
The heterodimeric ligand 9 containing (-)-1b and the (+)-
2b pharmacophores was prepared from sebacoyl monobenzyl
ester11 to yield the intermediate 7, which, after catalytic
hydrogenolysis, gave the corresponding acid 8.12 Treatment of
8 with the corresponding (+)-isomer (2b) in the presence of
DCC and DMAP13 yielded the heterodimeric ligand 9 (Scheme
2).
Similarly prepared were the heterodimeric ligands (10 and
11) by condensing the acid 8 with either cyclorphan (1c) or
morphine (1g). The ligand 12 was synthesized to evaluate the
effect of the spacer moiety on the binding affinity of these
compounds. Compound 13 with an ester linkage at one end of
the linking chain and an amide linkage at the other end was
obtained by coupling acid 8 with the 3-aminomorphinan 1f,
which was prepared from levorphanol (1a) by a previously
described procedure.14 The homodiamides 14, 15, and 16 were
prepared from 3-aminomorphinan (1f) in the presence of
succinyl chloride, sebacoyl chloride, or fumaryl chloride in the
presence of triethylamine respectively (Scheme 3).
To further investigate such opioid bivalent ligands containing
pharmacophores that have established κ/µ affinity, we prepared
* To whom correspondence should be addressed. Phone: 617-855-3388.
Fax: 617-855-2519. E-mail: Neumeyer@mclean.harvard.edu.
† Harvard Medical School.
‡ University of Rochester.
10.1021/jm050577x CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/03/2005

Homo- and Heterodimeric Ligands with Pharmacophores
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1 257
Table 1. Ki Values for the Inhibition of µ, δ, and κ Opioid Binding to CHO Membranes by Mono- and Bivalent Ligands
a Reference 8. b Reference 9. c Reference 1. d Reference 14.

258 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1
Peng et al.
cophores with its (+)-enantiomer as in 9 should lead to a
potency decrease due to the lower affinity of the (+)-enantiomer
at the neighboring opioid receptor. The affinity of the (+)-
enantiomer 2b is around 700 times lower than that of the
corresponding (-)-isomer (butorphan, 1b) at the κ receptor and
around 260 times lower at the µ receptor. Thus, if the (+)-
enantiomer interacts nonspecifically with the vicinal opioid site
or other membrane sites, the affinity of the (+),(-)-ligand 9
should be lower than that of the homoligand containing only
levo pharmacophores but more than that of the monovalent
ligands (8 or 12). The (+),(-)-ligand 9 contained the same
spacer length as the (-),(-)-bivalent ligand 18 that afforded
peak activity at the κ receptor. It was observed that the
heteroisomer possess 108-fold greater affinity as the (+),(+)-
ligand 3, but 24 times less than that of the (-),(-)-ligand 18
(Table 1). The finding that the (+),(-)-ligand 9 possesses lower
affinity at the κ receptor than the monomers 8 and 12 is possibly
due to the occupation of a neighboring site by (+)-morphinan,
which precludes access by another pharmacophore. Similarly,
because the (-),(-)-ligand 18 exhibits a potency that is greater
than that derived from the sum of its two monovalent pharma-
cophors, it can be considered a bivalent ligand that could bridge
two opioid receptors.
Figure 1. Morphinan monomer pharmacophores.
Scheme 1
At this point, we still could not establish if the second
pharmacophore itself or a specific moiety in the second
pharmacophore is responsible for the affinity. We thus continued
our investigation by synthesizing the butorphan-derived ligands
containing other pharmacophores. Initially, amide linkages were
introduced, as in 13, with an ester linkage at one end and an
amide at the other, and ligand 15, with amide linkage at both
ends. Both ligands (13 and 15) were found to have lower
affinities than the diester of butorphan 18 (by 7.7- and 153-
fold, respectively, for κ). It can thus be assumed that an ester
linkage is important for affinity. Heterodimeric compounds such
as 10 (1b combined with cyclorphan) and 11 (1b combined with
morphine) with a C-10 linking ester chain have similar high
affinitities (Ki ) 0.23-0.11 nM for κ and 0.34-0.22 nM for µ
receptor) to the monovalent ligand (8 or 12) but have lower
affinity than the (-),(-)-ligand 18. Compared to the (-),(-)-
ligand 18, compounds 4 [with N-[(S)-tetrahydrofurfuryl]mor-
phinan as a pharmacophore] and 5 (with N-propargylmorphinan
as a pharmacophore), differing only in their N-substituents on
the morphinan moiety, decrease the affinity by 24- and 204-
fold at the κ receptor. If hydrolysis of such esters in compound
18 had occurred, compound 4, 5, and 10 would have higher
affinity at µ and κ receptors than 18 because their monomeric
forms are more potent than butorphan (1b). We thus concluded
that the affinities of the (-),(-)-ligand 18 was not due to the
hydrolysis of the ester moieties resulting in two monomeric
ligands producing butorphan (1b) but rather can be ascribed to
the interaction of the dimeric ligand 18 to a ligand that could
bridge two opioid receptor sites.
Pharmacological Results and Discussion
Affinity and Selectivity of the Synthesized Ligands. The
dimeric and bivalent ligands that were investigated contain levo
pharmacophores (see Figure 1) that confer good agonist
activities or their less active dextro enantiomers (2b and 2c).
The spacer length for these compounds was dictated by the peak
potency that was observed when a sebacoyl ester (10 carbon)
unit was incorporated into the molecule. All the novel dimeric
morphinan ligands were evaluated for their affinity at and
selectivity for µ, δ, and κ opioid receptors with Chinese hamster
ovary (CHO) cell membranes stably expressing the human
opioid receptors. The data are summarized in Table 1. For
comparison purposes, opioid binding affinity data for 1a-f and
the enantiomers of 1b and 1c (i.e., 2b and 2c) are included in
Table 1.
Initially, the monovalent ligands were synthesized as controls
in order to evaluate the contribution of the spacer itself to
binding. The structures of compound 12 and 8 are similar, both
contain 1b (butorphan) as a pharmacophore with a C-10 ester
linking chain. They both have lower affinity than 1b. Since the
interaction between the 3-phenolic hydroxyl group and opioid
receptor proteins was historically recognized as crucial for high
affinity, it is reasonable that these ligands with a 3-position ester
spacer possess lower affinity than butorphan.
In our earlier report,1 a bivalent ligand containing two
identical (-)-morphinan pharmacophores with a 10-carbon
spacer (18) was the most potent ligand in the series, displaying
excellent affinities at µ and κ receptors. However, since it is
not known whether such enhancements reflected bridging of
opioid recognition sites on a receptor dimer, binding to an
accessory site, or results from the hydrolysis of the dimer, we
have attempted to distinguish between these possibilities by the
synthesis and evaluation of the affinities of dimers with other
pharmacophores.
When comparisons are made with the (-),(-)-ligand 18 (1b
with a 10-carbon chain linker), it is the second pharmacophore
itself, not a specific substituent in the second pharmacophore,
that is responsible for the κ affinity. Other structural modifica-
tions, such as an amide linkage (i.e., 13 and 15), (+)-enantiomer
(9), morphine (11), and N-substituent other than cyclobutyl
methyl (10, 4, 5), all reduce the κ affinity.
The dimeric diamides 14 and 16 were made to be compared
with the conformationally constrained fumaryl ester 19,1,2 which
is among the most potent ligands synthesized, displaying
excellent affinities at µ and κ receptors. When two amide
linkages were introduced with a fumaryl linkage as in 16, the
Assuming that such dimeric ligands bridge vicinal opioid
receptors, replacement of one of the (-)-morphinan pharma-

Homo- and Heterodimeric Ligands with Pharmacophores
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1 259
Scheme 2
Table 2. Agonist and Antagonist Properties of Compounds in
Scheme 3
Stimulating [35S]GTPγS Binding Mediated by the κ Opioid Receptora
pharmacological
compound
properties
Emax (%) EC50 (nM)
110 ( 2.0 46 ( 16
90 ( 10 0.19 ( 0.04
80 ( 6.8 1.3 ( 0.4
140 ( 4.1 8.8 ( 2.1
Imax (%)
(-)-U50,488
1c (cyclorphan) agonist
1b (butorphan)
agonist
-
-
agonist
agonist
agonist
agonist
agonist
agonist
agonist
agonist
-
4
6
8
9
10
11
12
13
no effect
no effect
no effect
no effect
no effect
no effect
no effect
65 ( 5.5
(at 10 µM)
91 ( 8.5
17 ( 7.6
75 ( 1.1 2.8 ( 0.4
77 ( 2.4 7.9 ( 0.3
110 ( 13
2.5 ( 0.5
150 ( 3.3 2.3 ( 0.3
170 ( 5.9 3.8 ( 0.2
55 ( 5.0 8.8 ( 2.6
partial agonist
a Membranes from CHO cells that stably expressed only the κ opioid
receptor were incubated with varying concentrations of the compounds.
The stimulation of [35S]GTPγS binding was measured as described in the
Experimental Section. To determine the antagonist properties of a compound,
membranes were incubated with 100 nM of the κ agonist U50,488 in the
presence of varying concentrations of the compound. The Emax value is the
maximal percent stimulation obtained with the compound. The EC50 value
is the concentration of compound needed to produce 50% of the Emax value.
The Imax value is the maximal percent inhibition obtained with the
compound. The IC50 value is the concentration of compound needed to
produce half-maximal inhibition. Dashed lines indicate that the compound
was not tested for antagonist properties because of its high Emax value.
affinity decreased by 100-fold. Similarly, compound 14 with a
succinyl amide linkage decreased the affinity by 50-fold
compared to the compound with the succinyl ester linking
butorphan 17. It is obvious that ester linkages are preferred to
amide linkages, as in 13 and 15. Compound 6 with the
conformationally constrained muconyl spacer was synthesized
and its affinity decreased by 7.9-fold at κ and 14-fold at µ.
Spacer length is thus a critical factor with respect to the ability
of the bivalent ligands to bridge neighboring receptors. A spacer
of insufficient length would not permit bridging, and an
excessively long spacer would tend to reduce bridging by
increasing the confinement volume of the free pharmacophore
so that it would spend less time in the vicinity of the unoccupied,
neighboring recognition site.
Efficacy Assay of Selected Ligands. To characterize the
relative efficacy of these morphinan ligands, compounds 4, 6,
and 8-13, together with cyclorphan (1c) and butorphan (1b),
were selected for the [35S]GTPγS assay. Table 2 shows the
agonist and antagonist properties of these ligands in stimulating
[35S]GTPγS binding mediated by the κ opioid receptor. Ligands
4, 10, 11, and 12 produced high maximal stimulation of [35S]-
GTPγS binding (Emax) comparable to that of selective agonist
U50,488, while ligands 6, 8, and 9 produced similar maximal
stimulation comparable to that of the monomers cylorphan (1c)
and butorphan (1b) but less than that of selective agonist U50,-
488. The EC50 values of these ligands are similar, which is
substantially correlated with the Ki values obtained for the
compounds in the binding assays with [3H]U69,593, except
ligand 6 had a somewhat higher EC50 value. Similar to the parent
compound butorphan (1b), most of these ligands except
compound 13 did not inhibit U50,488-stimulated [35S]GTPγS,
which suggested that most of these ligands were fully selective
κ agonists. Ligand 13 produced modest maximal stimulation
of [35S]GTPγS binding and also had a sound maximal inhibition
(Imax) of the U50,488-stimulated [35S]GTPγS binding. These
data indicated that ligand 13 was a partial κ agonist.The agonist
and antagonist properties of these ligands in stimulating [35S]-
GTPγS binding mediated by the µ opioid receptor are shown
in Table 3. Ligands 6, 8, 11, and 12 produced high maximal
stimulation of [35S]GTPγS binding mediated by the µ receptor
comparable to that of the full opioid agonist DAMGO. Ligands
4, 10, and 13 not only produced modest maximal stimulation
of [35S]GTPγS binding (Emax) but also produced sound maximal
inhibition (Imax) of the DAMGO-stimulated [35S]GTPγS binding.
These data indicated that ligands 4, 10, and 13 were µ agonists

260 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1
Peng et al.
Table 3. Agonist and Antagonist Properties of Compounds in Stimulating [35S]GTPγS Binding Mediated by the µ Opioid Receptora
pharmacological
compound
DAMGO
1c (cyclorphan)
1b (butorphan)
4
6
8
9
10
11
12
13
properties
Emax (%)
EC50 (nM)
Imax (%)
IC50 (nM)
agonist
120 ( 12
40 ( 2.9
50 ( 2.5
69 ( 3
96 ( 13
110 ( 8
28
41 ( 2
110 ( 11
93 ( 9
56 ( 6
110 ( 9.0
0.80 ( 0.06
1.6 ( 0.2
12 ( 1
-
-
partial agonist
partial agonist
agonist/antagonist
agonist
50 ( 1
50 ( 3
48 ( 4
no effect
no effect
71 ( 2
64 ( 2
no effect
no effect
61 ( 1
1.7 ( 0.4
20 ( 3
360 ( 86
no effect
no effect
460 ( 43
96 ( 21
no effect
no effect
150 ( 76
19 ( 2
agonist
3.0 ( 0.6
agonist/antagonist
partial agonist
agonist
agonist
partial agonist
NA
2.9 ( 0.8
2.5 ( 0.4
5.1 ( 1.3
4.4 ( 2.0
a Membranes from CHO cells that stably expressed only the µ opioid receptor were incubated with varying concentrations of the compounds. The stimulation
of [35S]GTPγS binding was measured as described in the Experimental Section. The Emax value is the maximal percent stimulation obtained with the
compound. The EC50 value is the concentration of compound needed to produce 50% of the Emax value. When the Emax value was 30% or lower, it was not
possible to calculate an EC50 value. To determine the antagonist properties of a compound, membranes were incubated with 200 nM of the µ agonist
DAMGO in the presence of varying concentrations of the compound. The Imax value is the maximal percent inhibition obtained with the compound. The IC50
value is the concentration of compound needed to produce half-maximal inhibition. Dashed lines indicate that the compound was not tested.
(621 mg) was dissolved in ethyl acetate (20 mL), and 10% Pd/C
(550 mg) was added. The air in the reaction bottle was flushed
with H2 three times. The solution was then hydrogenated at room
temperature overnight with Pd/C after which the catalyst was
removed by filtration. The solvent was evaporated to yield 450 mg
and antagonists. Ligand 9 had lowest maximal stimulation of
[35S]GTPγS binding and highest maximal inhibition (Imax) of
the DAMGO-stimulated [35S]GTPγS binding, indicating that
ligand 9 is a µ antagonist and weak agonist.
The preliminary assay for agonist and antagonist properties
of these ligands in stimulating [35S]GTPγS binding mediated
by the κ opioid receptor illustrated that most of these ligands
were κ agonists or partial agonists. But for the µ receptor,
compounds 6, 8, 11, and 12 were agonists, while compounds
4, 9, 10, and 13 were µ agonist/antagonist or partial agonists.
1
(85.6%) of a colorless oil, 8. H NMR (CDCl3, 400 MHz): 7.10
(d, J ) 8.8 Hz, 1H), 6.92 (d, J ) 1.6 Hz, 1H), 6.85 (dd, J ) 8.4,
1.2 Hz, 1H), 3.47 (s, 1H), 3.01 (d, J ) 18.4 Hz, 1H), 2.81 (s, 1H),
2.62-1.04 (m, 39H). 13C NMR (CDCl3, 75 MHz): 178.5, 172.3,
150.2, 139.8, 130.9, 128.9, 119.9, 118.6, 58.7, 56.2, 45.9, 40.6,
38.4, 36.5, 35.4, 35.2, 34.3, 31.1, 29.2, 29.2, 29.1, 29.0, 27.9, 27.7,
25.9, 25.6, 25.4, 24.8, 24.5, 21.6, 18.6. Anal. (C31H45NO4
.0.5H2O.1.5HCl) C, H, N.
Experimental Section
General Procedure for the Preparation of Ligands 9-13.
Method A. The acid 8 (0.6 mmol) and an appropriate opioid (0.5
mmol) was dissolved in anhydrous dichloromethane (15 mL) under
nitrogen. A catalytic amount of 4-(dimethylamino)pyridine was
added, followed by N,N′-dicyclohexylcarbodiimide (0.6 mmol). The
solution mixture was stirred at room temperature overnight, the
solid was filtered off, and the crude product was purified by column
chromatography on silica gel (EtOAc:Et3N, 100:1) to afford the
corresponding bivalent ligands.
Method B. The acid 8 (0.5 mmol) was dissolved in anhydrous
dichloromethane (25 mL), cooled to 0°C, and oxalyl chloride (2
mmol) and 2 drops of DMF were added. The mixture was stirred
at 0 °C to room temperature for 2 h and then evaporated to dryness,
and excess of oxalyl chloride was removed under high vacuum.
The acid chloride formed was dissolved in dichloromethane (25
mL) and cooled, to which the appropriate opioid (0.5 mmol) was
added, followed by triethylamine (4 mmol). The reaction mixture
was stirred at 0 °C-25 °C overnight. Then saturated aqueous
NaHCO3 solution (20 mL) was added, and the mixture was
extracted with EtOAc (3 × 20 mL) and washed with brine. The
extract was evaporated to dryness under high vacuum. The crude
product was purified by column chromatography on silica gel
(EtOAc:Et3N, 100:1) to afford the corresponding bivalent ligands.
1-((-)-N-Cyclobutylmethylmorphinan-3-yl) 10-((+)-N-cy-
clobutylmethylmorphinan-3-yl) sebacoylate (9): light yellow oil
(73.1%). 1H NMR (300 MHz): 7.10 (d, J ) 8.4 Hz, 2H), 6.92 (d,
J ) 2.4 Hz, 2H), 6.85 (dd, J ) 8.1, 2.1 Hz, 2H), 3.01 (d, J ) 18.6
Hz, 2H), 2.82-2.79 (m, 2H), 2.62-1.23 (m, 62H). 13C NMR (75
MHz):172.3, 149.2, 142.0, 135.2, 128.4, 118.4, 118.0, 61.4, 55.8,
45.6, 44.9, 41.7, 37.7, 36.5, 34.9, 34.3, 29.0, 28.9, 27.7, 27.7, 26.7,
26.5, 24.8, 24.3, 22.0, 18.8. Anal. (C52H72N2O5‚1.5H2O) C, H, N
1-((-)-N-Cyclopropylmethylmorphinan-3-yl) 10-(N-cyclobu-
tylmethylmorphinan-3-yl) sebacoylate (10): light yellow oil
(80.9%). 1H NMR (300 MHz): 7.10 (d, J ) 8.1 Hz, 2H), 6.93 (d,
J ) 2.1 Hz, 2H), 6.85 (dd, J ) 8.1, 2.1 Hz, 2H), 3.102-3.017 (d,
J ) 18.6 Hz, 2H), 2.956-2.895 (d, J ) 18.3 Hz, 2H), 2.82-1.00
(m, 60H). 13C NMR (75 MHz): 172.4, 149.1,149.2, 142.0, 135.2,
Melting points were determined on a Thomas-Hoover capillary
1
tube apparatus and are reported uncorrected. H and 13C NMR
spectra were recorded on a Bruker AC300 spectrometer using
tetramethylsilane as an internal reference. Element analyses,
performed by Atlantic Microlabs, Atlanta, GA, were within 0.4%
of theoretical values. Analytical thin-layer chromatography (TLC)
was carried out on 0.2 mm Kieselgel 60F 254 silica gel plastic
sheets (EM Science, Newark). Flash chromatography was used for
the routine purification of reaction products. The column output
was monitored by TLC.
(+)-3-Hydroxy-N-cyclopropylmethylmorphinan Mandelate
(2c). (+)-3-Hydroxy-N-cyclobutylmethylmorphinan free base was
made from commercially available dextrorphan (Sigma), which was
converted to the S-(+)-mandelate salt and recrystallized from
isopropyl ether/EtOH. The product is obtained as a white solid,
mp 204-205 °C. Anal. (C20H27NO‚C8H8O3) C, H, N
(+)-3-Hydroxy-N-cyclobutylmethylmorphinan mandelate (2b)
was similarly prepared from dextrophan and converted to the (S)-
(+)-mandelate salt to yield a light brown solid, mp 208-210 °C.
Anal. (C21H29NO‚C8H8O3‚0.2H2O) C, H, N
Sebacic Acid 10-((-)-N-cyclobutylmethylmorphinan-3-yl) es-
ter sebacic acid (8). The monobenzyl ester of sebacic acid (1.7
mmol) was dissolved in anhydrous dichloromethane (25 mL) and
cooled to 0 °C, and oxalyl chloride (500 µL) and 2 drops of DMF
were added. The mixture was stirred at 0 °C to room temperature
for 2 h and then evaporated to dryness, and the excess of oxalyl
chloride was removed under high vacuum. The acid chloride was
dissolved in dichloromethane (25 mL) and cooled to 0 °C, and
butorphan 1b (361 mg, 1.1 mmol) was added. Then triethylamine
(0.74 mL) was added dropwise. The reaction mixture was stirred
at 0 °C to room temperature overnight, saturated aqueous NaHCO3
solution (30 mL) was added, and the reaction mixture was extracted
with EtOAc (3 × 30 mL) and washed with brine. The EtOAc extract
was evaporated to dryness under high vacuum. The residual oil
was purified by flash chromatography (SiO2, EtOAc:TEA, 100:5)
to afford a colorless, oily compound (671.0 mg, 90.7%). The 3-(10-
benzyl ester) sebacic acid ester N-(cyclobutylmethyl)morphinan 7

Homo- and Heterodimeric Ligands with Pharmacophores
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1 261
135.2, 128.4, 118.4, 118.0, 61.5, 60.0, 55.8, 55.6, 46.2, 45.6, 45.6,
44.9, 41.7, 37.9, 37.7, 36.5, 34.9, 34.4, 32.7, 30.8, 29.0, 29.0, 27.8,
26.7, 26.5, 26.4, 24.8, 24.7, 24.3, 24.2, 22.1, 18.8, 11.5, 9.4, 4.03,
3.4. Anal. (C51H70N2O4‚0.25Et3N‚0.25H2O) C, H, N.
Bis((-)-N-cyclobutylmethylmorphinan-3-yl) muconate (6):
yellow solid (70.6%). Mp: 79-81 °C. 1H NMR (CDCl3, 400
MHz): 7.59 (dd, J ) 11.6, 3.2 Hz, 2H), 7.14 (d, J ) 8.0 Hz, 2H),
7.00 (d, J ) 2.4 Hz, 2H), 6.92 (dd, J ) 8.0, 2.4 Hz, 2H), 6.46 (dd,
J ) 11.6, 3.2 Hz, 2H), 3.03 (d, J ) 18.4 Hz, 2H), 2.83 (m, 2H),
2.63-0.88 (m, 46H). 13C NMR (CDCl3, 75 MHz): 164.3, 148.9,
142.3, 141.9, 135.7, 128.6, 128.4, 118.3, 117.9, 61.4, 55.8, 45.6,
44.8, 41.7, 37.8, 36.5, 34.9, 27.8, 26.7, 26.5, 24.4, 22.1, 18.8. Anal.
(C48H60N2O4‚H2O) C, H, N.
1-((-)-N-Cyclobutylmethylmorphinan-3-yl) 10-(3,6-diol-7,8-
didehydro-4,5-epoxy-17-methyl-(5r,6r)morphinan-3-yl) seba-
coylate (11): light yellow oil (62.2%). 1H NMR (CDCl3, 300
MHz): 7.10 (d, J ) 8.1 Hz, 1H), 6.92 (d, J ) 2.4 Hz, 1H), 6.85
(dd, J ) 8.1, 2.4 Hz, 1H), 6.74 (d, J ) 8.1 Hz, 1H), 6.61 (d, J )
8.1 Hz, 1H), 5.77 (d, J ) 9.9 Hz, 1H), 5.30 (dt, J ) 9.9, 2.7 Hz,
1H), 4.93 (dd, J ) 6.9, 0.9 Hz, 1H), 4.16-1.24 (m, 56H). 13C NMR
(CDCl3, 75 MHz): 172.4, 171.5, 149.2, 148.7, 142.0, 135.2, 134.1,
132.7, 132.2, 131.8, 128.4, 127.7, 121.0, 119.9, 118.4, 118.1, 92.2,
65.7, 61.4, 58.8, 55.7, 46.3, 44.9, 43.0, 42.5, 41.7, 40.4, 37.7, 36.5,
35.2, 34.9, 34.4, 33.9, 29.0, 28.8, 27.8, 26.7, 26.5, 24.8, 24.3, 22.1,
20.6, 18.8. Anal. (C48H64Cl2N2O6‚0.5H2O) C, H, N.
1-Phenyl 10-((-)-N-cyclobutylmethylmorphinan-3-yl) seba-
coylate (12): light yellow oil (72.2%); 1H NMR (CDCl3, 400
MHz): 7.39 (s, 5H), 7.10 (d, J ) 8.4 Hz, 1H), 6.92 (d, J ) 1.6
Hz, 1H), 6.85 (dd, J ) 8.0, 2.4 Hz, 1H), 5.12 (s, 2H), 3.01 (d, J )
18.4 Hz, 1H), 2.86 (s, 1H), 2.65-1.06 (m, 37H). 13C NMR (CDCl3,
100 MHz): 172.3, 172.2, 150.7, 149.2, 142.0, 135.1, 129.3, 128.4,
125.6, 121.5, 118.5, 118.1, 61.4, 55.8, 45.6, 44.8, 41.6, 37.7, 36.5,
34.8, 34.3, 34.3, 29.0, 29.0, 27.8, 26.7, 26.4, 24.8, 24.4. Anal.
(C37H49NO4‚0.2H2O) C, H, N.
Bis(3-amino-(-)-N-cyclobutylmethylmorphinan) succinamide
(14): off-white solid (61.2%). Mp: 165-167 °C. 1H NMR (CDCl3,
300 MHz): 7.47 (d, J ) 2.0 Hz, 2H), 7.34 (dd, J ) 8.4, 2.0 Hz,
2H), 7.06 (d, J ) 8.0 Hz, 2H), 4.90 (s, 2H), 3.31-3.29 (m, 2H),
3.04 (d, J ) 19.2 Hz, 2H), 2.94 (s, 2H), 2.72-1.06 (m, 48H). 13
C
NMR (CDCl3, 75 MHz): 172.9, 141.3, 138.5, 133.8, 129.2, 119.2,
118.2, 62.0, 57.4, 47.2, 45.3, 42.0, 38.5, 37.4, 35.2, 32.7, 29.0, 28.8,
27.9, 27.6, 25.2, 23.3, 19.6. Anal. (C46H62N4O2‚1.5H2O) C, H, N.
Bis(3-amino-(-)-N-cyclobutylmethylmorphinan) sebacamide
1
(15): white solid (58.8%). Mp: 165-167 °C. H NMR (CDCl3,
300 MHz): 7.61 (s, 2H), 7.36 (d, J ) 6.3 Hz, 4H), 7.03 (d, J )
8.4 Hz, 2H), 2.98 (d, J ) 18.6 Hz, 2H), 2.79 (s, 2H), 2.59-1.24
(m, 62H). 13C NMR (CDCl3, 75 MHz): 171.4, 141.2, 136.3, 133.8,
128.0, 117.4, 116.6, 61.5, 55.8, 45.7, 45.04, 41.8, 37.6, 37.6, 36.5,
34.8, 29.0, 28.9, 27.8, 26.8, 26.6, 25.4, 24.4, 22.2, 18.8. Anal.
(C52H74N4O2‚0.5H2O) C, H, N.
1-((-)-N-Cyclobutylmethylmorphinan-3-yl) 10-((-)-N-cy-
clobutylmethylmorphinan-3-yl) sebacoylate (13): light yellow
oil (42.7%); 1H NMR (300 MHz): 7.35-7.28 (m, 3H), 7.06 (d, J
) 8.4 Hz, 1H), 7.01 (d, J ) 8.0 Hz, 1H), 6.88 (d, J ) 2.4 Hz, 1H),
6.81 (dd, J ) 8.0, 2.0 Hz, 1H), 2.97 (dd, J ) 18.8, 10.4 Hz, 2H),
2.77 (d, J ) 2.8 Hz, 2H), 2.57-1.02 (m, 62H). 13C NMR (75 MHz):
172.4, 171.2, 149.2, 142.0, 141.3, 136.2, 135.3, 133.9, 128.4, 128.0,
118.4, 118.0, 117.4, 116.6, 61.5, 60.3, 55.8, 55.7, 45.7, 45.6, 45.1,
44.9, 41.9, 41.7, 37.7, 37.6, 36.5, 34.9, 34.3, 33.9, 29.14, 29.09,
28.99, 28.98, 27.79, 26.8, 26.7, 26.5, 26.5, 25.6, 25.5, 24.9, 24.8,
24.4, 22.2, 22.1, 18.8, 14.1. Anal. (C52H73N3O3‚0.25H2O) C, H, N.
General Procedure for the Preparation of Ligands 3-6, 14-
16. To a cooled solution of the appropriate morphinan (1 mmol)
and Et3N (0.5 mL) in anhydrous CH2Cl2 (10 mL), the appropriate
acid dichloride (90.6 mmol) was added dropwise at 0 °C. The
solution was stirred at room temperature for 48 h and then diluted
with dichloromethane. The organic layer was separated, washed
with 10% NaHCO3 and brine, dried with Na2SO4, and then
concentrated to yield a dark oil. The crude product was purified
by column chromatography on silica gel (EtOAc:Et3N, 100:1) to
afford the corresponding bivalent ligands.
Bis((-)-N-cyclobutylmethylmorphinan-3-amino) fumaramide
(16): light-brown solid (50.5%). Anal. (C46H60N4O2‚0.75H2O) C,
H, N.
Opioid Binding to the Human µ, δ, and K Opioid Receptors.
Chinese hamster ovary (CHO) cells stably transfected with the
human κ opioid receptor (hKOR-CHO), δ opioid receptor (hDOR-
CHO), and the µ opioid receptor (hMOR-CHO) were obtained from
Drs. Larry Toll (SRI International, Palo Alto, CA) and George Uhl
(NIDA Intramural Program, Bethesda, MD). The cells were grown
in 100 mm dishes in Dulbecco’s modified Eagle’s media (DMEM)
supplemented with 10% fetal bovine serum (FBS) and penicillin-
streptomycin (10 000 units/mL) at 37 °C in a 5% CO2 atmosphere.
The affinity and selectivity of the compounds for the multiple opioid
receptors were determined by incubating the membranes with
radiolabeled ligands and 12 different concentrations of the com-
pounds at 25 °C in a final volume of 1 mL of 50 mM Tris-HCl,
pH 7.5. Incubation times of 60 min were used for the µ-selective
peptide [3H]DAMGO and the κ-selective ligand [3H]U69,593. A
3-h incubation was used with the δ-selective antagonist [3H]-
naltrindole.
[35S]GTPγS Binding Studies To Measure Coupling to G
Proteins. Membranes from CHO cells stably expressing either the
human κ or µ opioid receptor were used in the experiments. Cells
were scraped from tissue culture plates and then centrifuged at
1000g for 10 min at 4 °C. The cells were resuspended in phosphate-
buffered saline, pH 7.4, containing 0.04% EDTA. After centrifuga-
tion at 1000g for 10 min at 4 °C, the cell pellet was resuspended
in membrane buffer, which consisted of 50 mM Tris-HCl, 3 mM
MgCl2, and 1 mM EGTA, pH 7.4. The membranes were homog-
enized by with a Dounce homogenizer, followed by centrifugation
at 40000g for 20 min at 4 °C. The membrane pellet was resuspended
in membrane buffer, and the centrifugation step was repeated. The
membranes were then resuspended in assay buffer, which consisted
of 50 mM Tris-HCl, 3 mM MgCl2, 100 mM NaCl, and 0.2 mM
EGTA, pH 7.4. The protein concentration was determined by the
Bradford assay using bovine serum albumin as the standard. The
membranes were frozen at -80 °C until use.
CHO cell membranes expressing either the human κ opioid
receptor (15 µg of protein per tube) or µ opioid receptor (7.5µg of
protein per tube) were incubated with 12 different concentrations
of the agonist in assay buffer for 60 min at 30 °C in a final volume
of 0.5 mL. The reaction mixture contained 3 µM GDP and 80 pmol
of [35S]GTPγS. Basal activity was determined in the presence of 3
µM GDP and in the absence of an agonist, and nonspecific binding
was determined in the presence of 10 µM unlabeled GTPγS. Then,
the membranes were filtered onto glass-fiber filters by vacuum
Bis((+)-N-cyclobutylmethylmorphinan-3-yl) sebacoylate (3):
1
colorless oil (69.7%). H NMR (CDCl3, 300 MHz): 7.10 (d, J )
8.1 Hz, 2H), 6.92 (d, J ) 2.1 Hz, 2H), 6.85 (dd, J ) 8.1, 2.1 Hz,
2H), 3.01 (d, J ) 18.6 Hz, 2H), 2.82-1.23 (m, 64H). 13C NMR
(CDCl3, 75 MHz): 172.4, 149.2, 142.1, 135.3, 128.4, 118.4, 118.1,
61.5, 55.8, 45.6, 44.9, 41.8, 37.8, 36.6, 34.9, 34.4, 29.1, 29.0, 27.8,
26.8, 26.5, 24.8, 24.4, 22.1, 18.8. Anal. (C52H72N2O4‚H2O) C, H,
N.
Bis((-)-N-[(S)-tetrahydrofurfuryl]morphinan-3-yl) sebacoy-
1
late (4): light yellow oil (81.4%). H NMR (CDCl3, 300 MHz):
7.10 (d, J ) 8.4 Hz, 2H), 6.92 (d, J ) 2.4 Hz, 2H), 6.85 (dd, J )
8.4, 2.4 Hz, 2H), 4.00 (m, 2H), 3.92 (m, 2H), 3.78 (m, 2H), 3.01-
1.05 (m, 60H). 13C NMR (CDCl3, 75 MHz): 172.3, 149.1, 142.0,
135.3, 128.4, 118.4, 118.0, 77.6, 68.0, 60.1, 56.8, 45.8, 44.7, 41.7,
37.6, 36.4, 34.3, 30.2, 29.0, 29.0, 26.6, 26.5, 25.4, 24.9, 24.8, 22.1.
Anal. (C52H72N2O6) C, H, N.
Bis((-)-N-propargylmorphinan-3-yl) sebacoylate (5): color-
1
less oil (63.2%). H NMR (CDCl3, 300 MHz): 7.10 (d, J ) 8.4
Hz, 2H), 6.94 (d, J ) 2.4 Hz, 2H), 6.86 (dd, J ) 8.4, 2.4 Hz, 2H),
3.41 (dt, J ) 9.9, 2.7 Hz, 4H), 3.14 (m, 2H), 3.02 (d, J ) 18.4 Hz,
2H), 2.70-2.62 (dt, J ) 9.9, 2.7 Hz, 4H), 2.56-2.52 (t, J ) 2.8
Hz, 4H), 2.32-1.07 (m, 38H). 13C NMR (CDCl3, 75 MHz): 172.3,
149.3, 141.7, 134.7, 128.5, 118.6, 118.1, 80.6, 72.3, 55.6, 45.2,
44.8, 43.9, 41.6, 37.5, 36.4, 34.3, 29.0, 28.9, 26.5, 26.4, 24.8, 24.3,
22.0. Anal. (C48H60N2O4) C, H, N.

262 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1
Peng et al.
Length and Selectivity at Multiple Opioid Receptors. J. Med. Chem.
1986, 29, 1855-1861. (d) Portoghese, P. S.; Nagase, H.; Takemori,
A. E. Only One Pharmacophore Is Required for the κ Opioid
Antagonist Selectivity of Norbinaltorphimine. J. Med. Chem. 1988,
31, 1344-1347. (e) Portoghese, P. S.; Ronsisvalle, G.; Larson, D.
L.; Takemori, A. E. Synthesis and Opioid Antagonist Potencies of
Naltrexamine Bivalent Ligands with Conformationally Restricted
Spacers. J. Med. Chem. 1986, 29, 1650-1653. (f) Portoghese, P. S.;
Larson, D. L.; Yim, C. B.; Sayre, L. M.; Ronsisvalle, G.; Lipkowski,
A. W.; Takemori, A. E.; Rice, K. C.; Tam, S. W. Stereostructure-
Activity Relationship of Opioid Agonist and Antagonist Bivalent
Ligands. Evidence for Bridging Between Vicinal Opioid Receptors.
J. Med. Chem. 1985, 28, 1140-1141.
filtration, followed by three washes with 3 mL of ice-cold 50 mM
Tris-HCl, pH 7.5. Samples were counted in 2 mL of Ecoscint A
scintillation fluid. Data represent the percent of agonist-stimulation
[35S]GTPγS binding over the basal activity, defined as [(specific
binding/basal binding) × 100] - 100. All experiments were
repeated at least three times and were performed in triplicate. To
determine antagonist activity of a compound at the µ opioid
receptors, CHO membranes expressing the µ opioid receptor were
incubated with the compound in the presence of 200 nM of the
agonist DAMGO. To determine antagonist activity of a compound
at the κ opioid receptors, CHO membranes expressing the κ opioid
receptor were incubated with the compound in the presence of 100
nM of the κ agonist U50,488.
(5) Portoghese, P. S. From Models to Molecules: Opioid Receptor
Dimers, Bivalent Ligands, and Selective Opioid Receptor Probes. J.
Med. Chem. 2001, 44, 2259-2269.
Acknowledgment. This work was supported in part by NIH
Grants RO1-DA14251 (J. L. N.) and K05-DA 00360 (J. M.
B.). Levorphanol tartrate was generously donated by Mallinck-
rodt Inc. We also wish to thank Drs. A. Zhang, S. Negus, and
N. Mello for helpful discussions.
(6) Daniels, J. D.; Kulkarni, A.; Xie, Z.; Bhushan, R. G.; Portoghese, P.
S. A Bivalent Lignd(KDAN-18) Containing δ-Antagonist and
κ-Agonist Pharmacophores Bridges δ2 and κ1 Opioid Receptor
Phenotypes. J. Med. Chem. 2005, 48, 1713-1716.
(7) Olfoson, R. A.; Marts, J. T.; Seret, J. P.; Piteau, M.; Malfroot, T. A.
A new Reagent for the Sevlective, High Yield N-Dealkylation of
Teriary Amines; Improved Synthesis of Naltrexone and Nalbuphine.
J. Org. Chem. 1984, 49, 2081-2082.
(8) Neumeyer, J. L.; Bidlack, J. M.; Zong, R.; Bakthavachalam, V.; Gao,
P.; Cohen, D. J.; Negus, S. S.; Mello, N. K. Synthesis and opioid
receptor affinity of morphinan and benzomorphan derivativessMixed
κ agonists and µ agonists/antagolnists as potential pharmacothera-
peutics for cocaine dependence. J. Med. Chem. 2000, 43, 114-122.
(9) Neumeyer, J. L.; Gu, X. H.; Van Vliet, L. A.; DeNunzio, N. J.;
Rusovici, D. E.; Cohen, D. J.; Negus, S. S.; Mello, N. K.; Bidlack,
J. M. Mixed κ agonists and µ agonists/antagonists as potential
pharmacotherapeutics for cocaine abuse: Synthesis and opioid
receptor binding affinity of N-substituted derivatives of morphinan.
Bioorg Med. Chem. Lett. 2001, 11, 3049-3053.
(10) (a) Bailey, D. S.; Ross, J. M. The Synthesis of 1,4-Dibenzoyl-1,3-
butadiene, J. Am. Chem. Soc. 1949, 71, 2370-2372. (b) Nicholas,
A. M.; Fles, D. Synthesis of Muconyldimalonates, J. Am. Chem. Soc.
1956, 78, 5903-6. (c) Tsuji, J.; Morikawa, M.; Iwamoto, N. Organic
Syntheses By Means of Noble Metal Compounds. VI. Synthesis of
Muconic Acid, J. Am. Chem. Soc. 1964, 86, 2095.
(11) Babler, J. H.; Moy, R. K. A facile method for monoesterification of
R,ω-dicarboxylic acids: Application to the synthesis of traumatic
acid, a prostaglandin synthon. Synth. Commun. 1979, 9, 669-75.
(12) Smith, H. L.; Brown, E. S.; Smith, J. D.; Andrako, J. Hydrogenolysis
of benzyl esters with palladium-on-carbon catalysts. J. Pharm. Sci.
1965, 54, 1269-1273.
(13) Andrade, C. K. Z.; Rocha, R. O.; Vercillo, O. E.; Silva, W. A.; Matos,
R. A. F. DCC/DMAP-mediated coupling of carboxylic acids with
oxazolidinones and thiazolidinethiones. Synlett. 2003, 15, 2351-
2352.
Supporting Information Available: Elemental analyses for all
final compounds listed in Table 1. This material is available free
of charge via the Internet at http://pubs.acs.org.
References
(1) Neumeyer, J. L.; Zhang, A.; Xiong, W.; Gu. X.; Hilbert, J. E.; Knapp,
B. I.; Negus, S. S.; Mello, N. K.; Bidlack, J. M. Design and Synthesis
of Novel Dimeric Morphinan Ligands for κ and µ Opioid Receptors.
J. Med. Chem. 2003, 46, 5162-5170.
(2) Mathews, J. L.; Peng, X.; Zhang, A.; Xiong, W.; Negus, S. S.;
Neumeyer, J. L.; Bidlack, J. M. Characterization of MCL-145: A
novel bivalent morphinan possessing mixed kappa agonist and mu
agonist/antagonist properties. J. Pharmacol. Exp. Ther. 2005, 315,
821-827.
(3) (a) Cvejic, S.; Devi, L. Dimerization of the δ Opioid Receptor:
Implication for a Role in Receptor Internalization. J. Biol. Chem.
1997, 272, 26959-26964. (b) Jordan, B. A.; Cvejic, S.; Devi, L. A.
Opioids and Their Complicated Receptor Complexes. Neuroopay-
chopharcacology 2000, 23, S15-S18. (c) George, S. R.; Fan, T.;
Xie, Z.; Tse, R.; Tamni, V.; Varghese, G.; O’Dowd, B. F. Oligo-
merization of µ- and δ-Opioid Receptors: Generation of Novel
Functional Properties. J. Biol. Chem. 2000, 275, 26128-26135.
(4) (a) Portoghese, P. S.; Ronsisvalle, G.; Larson, D. L.; Yim, C. B.;
Sayre, L. M.; Takemori, A. E. Opioid agonist and antagonist bivalent
ligands as receptor probes. Life Sci. 1982, 31, 1283. (b) Erez, M.;
Takemori, A. E.; Portoghese, P. S. Narcotic antagonistic potency of
bivalent ligands which contain beta-naltrexamine. Evidence for
bridging between proximal recognition sites. J. Med. Chem. 1982,
25, 847-9. (c) Portoghese, P. S.; Larson, D. L.; Sayre, L. M.; Yim,
C. B.; Ronsisvalle, G.; Tam, S. W.; Takemori, A. E. Opioid Agonist
and Antagonist Bivalent Ligands. The Relationship Between Spacer
(14) Zhang, A.; Vliet, A. V.; Neumeyer, J. L. Synthesis of Aminothiazole
Derived Morphinans. Tetrahedron Lett. 2003, 44, 6459-6462.
JM050577X