De novo design, synthesis, and biological activities of high-affinity and selective non-peptide agonists of the δ-opioid receptor

Article abstract of DOI:10.1021/jm980374r

On the basis of the structure-activity relationships of δ-opioid- selective peptide ligands and on a model of the proposed bioactive conformation for a potent and selective, conformationally constrained δ- opioid peptide ligand [(2S,3R)-TMT¹]DP

Full text of DOI:10.1021/jm980374r

J . Med. Chem. 1998, 41, 4767-4776
4767
De Novo Design , Syn th esis, a n d Biologica l Activities of High -Affin ity a n d
Selective Non -P ep tid e Agon ists of th e δ-Op ioid Recep tor |
Subo Liao,†,§ J osue Alfaro-Lopez,† Mark D. Shenderovich,†,∇ Keiko Hosohata,‡ J un Lin,† Xiaoping Li,‡
Dagmar Stropova,‡ Peg Davis,‡ Kevin A. J ernigan,‡ Frank Porreca,‡ Henry I. Yamamura,‡ and Victor J . Hruby*,†
Departments of Chemistry and Pharmacology, The University of Arizona, Tucson, Arizona 85721
Received J une 19, 1998
On the basis of the structure-activity relationships of δ-opioid-selective peptide ligands and
on a model of the proposed bioactive conformation for a potent and selective, conformationally
constrained δ-opioid peptide ligand [(2S,3R)-TMT1]DPDPE, a series of small organic peptide
mimetic compounds targeted for the δ-opioid receptor have been designed, synthesized, and
evaluated in radiolabeled ligand binding assays and in vitro bioassays. The new non-peptide
ligands use piperazine as a template to present the most important pharmacophore groups,
including phenol and phenyl groups and a hydrophobic moiety. This hydrophobic group was
designed to mimic the hydrophobic character of the D-Pen residues in DPDPE, which has been
found to be extremely important for increasing the binding affinity and selectivity of these
non-peptide ligands for the δ-opioid receptor over the µ-opioid receptor. Compound 6f (SL-
3111) showed 8 nM binding affinity and over 2000-fold selectivity for the δ-opioid receptor
over the µ-opioid receptor. Both enantiomers of SL-3111 were separated, and the (-)-isomer
was shown to be the compound with the highest affinity for the δ-opioid receptor found in our
study (IC50 ) 4.1 nM), with a selectivity very similar to that observed for the racemic compound.
The phenol hydroxyl group of SL-3111 turned out to be essential to maintain high affinity for
the δ-opioid receptor, which also was observed in the case of the δ-opioid-selective peptide
ligand DPDPE. Binding studies of SL-3111 and [p-ClPhe4]DPDPE on the cloned wild-type
and mutated human δ-opioid receptors suggested that the new non-peptide ligand has a binding
profile similar to that of DPDPE but different from that of (+)-4-[((RR)-R(2S,5R)-4-allyl-2,5-
dimethyl-1-piperazinyl)-3-methoxybenzyl]-N,N-diethylbenzamide (SNC-80), another δ-opioid-
selective non-peptide ligand.
In tr od u ction
required. On the basis of this information, the geo-
metrical relationships of key phamacophore elements
may become more clear.2 In the field of opioid research,
selective peptide agonists for the δ-opioid receptor have
shown promising therapeutic potential as analgesics
without the adverse side effects associated with mor-
phine and other opioid drugs selective for the µ-opioid
receptor.3 Detailed structure-activity relationship (SAR)
studies of opioid peptides have provided much informa-
tion about the chemical, conformational, dynamic, and
structural requirements for the selective agonist recog-
nition of the δ-opioid receptors.4 However, only a few
successful attempts have been reported in the develop-
ment of the δ-opioid agonist ligands of a non-peptidic
nature. Several examples of non-peptide ligands have
been discovered either by modification of morphine-type
alkaloids or by random screening approaches,5-7 but
most of them are associated, as potential drug candi-
dates, with various problems such as poor selectivity6
and low efficacy in vivo7 and display binding profiles
different from those of typical δ-opioid-selective peptide
agonists.8 Therefore, development of δ-opioid receptor-
selective non-peptide agonists, based on peptide struc-
ture-activity relationships, would have an important
impact in understanding relationships between peptide
and non-peptide ligand-receptor recognition, and signal
transduction requirements, for rational drug design and
for discovery of novel therapeutics.
A major but difficult goal of peptidomimetic design
is the translation of three-dimensional (3D) pharma-
cophore information from a potent bioactive peptide to
a small organic non-peptide ligand with the same
agonist or antagonist biological properties.1 Some
classes of non-peptide ligands may have advantages as
drug candidates, such as high oral bioavailability and
higher permeability across the blood-brain barrier. To
successfully bridge the gap from peptide to non-peptide
structures, a detailed development of a 3D peptide
pharmacophore model based on the design, synthesis,
and comprehensive evaluation of conformational (φ,ψ)
and topographical (chi[ø]) space of a peptide ligand is
| Abbreviations used for amino acids follow the rules of the IUPAC-
IUB J oint Commission of Biochemical Nomenclature (Biochem. J .
1984, 219, 345-373). Other abbreviations: TMT, â-methyl-2′,6′-
dimethyltyrosine; D-Pen, D-â,â-dimethylcysteine; DPDPE, c[D-Pen2,D-
Pen5]enkephalin; Pen, penicillamine; Aib, R-aminoisobutyric acid; ICI-
174,864, (allyl)2-Tyr-Aib-Aib-Phe-Leu-OH; SIOM, 7-spiroindanyloxy-
morphone; Boc, tert-butyloxycarbonyl; p-MeBzl, p-methylbenzyl; FAB-
MS, fast atom bombardment mass spectrometry; TLC, thin-layer
chromatography; p-MBHA, p-methylbenzhydrylamine; m-NBA, m-
nitrobenzyl alcohol; DCM, dichloromethane; DIEA, diisopropylethyl-
amine.
* To whom correspondence should be addressed.
† Department of Chemistry.
‡ Department of Pharmacology.
§ Current address: Monsanto Co., 800 N. Lindberg Blvd., St. Louis,
MO 63167.
∇ Current address: Structural Bioinformatics, Inc., 10929 Technol-
ogy Place, San Diego, CA, 92127.
10.1021/jm980374r CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/04/1998

4768 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Liao et al.
F igu r e 1. Structures of some typical δ-opioid-selective peptide and non-peptide ligands.
As part of our efforts in opioid ligand design, we are
systematically examining approaches to cross the gap
between peptide and non-peptide compounds selective
for the δ-opioid receptor.9 Recently, we have developed
the conformationally and topographically constrained
highly potent δ-opioid-selective peptide [(2S,3R)-TMT1]-
DPDPE.10 On the basis of the model of the bioactive
conformation proposed for this peptide lead,11 we have
designed and synthesized a series of non-peptide ligands
using the piperazine ring as a template. The lead
compound in this first generation of peptidomimetics
(SL-3111) shows high binding affinity, exceptionally
high selectivity for δ-opioid receptors, and a binding
profile similar to that of the lead peptide agonist
[(2S,3R)-TMT1]DPDPE. In this paper, we report the
design, synthesis, and pharmacological evaluation of
these novel δ-opioid-selective non-peptide agonists.
the blood-brain barrier.13 Most importantly, it was
found to produce a potent, efficacious, and prolonged
analgesia in several in vivo tests and to have low
addiction potential, making it an alternative candidate
for the treatment of pain without the toxicity and side
effects of current drugs. Comprehensive 2D NMR,4b
computational studies,14 and a recently determined
X-ray crystal structure of DPDPE15 have led to the
conclusion that the 14-membered ring maintains a very
similar conformation in solution and in the crystal form.
However, these studies also showed that the side-chain
groups of the important Phe4 and Tyr1 phamacophores
were highly flexible. Therefore, further constraints
were applied to the side chains of these residues in
DPDPE, using novel topographically constrained amino
acids. In one of these studies, the substitution of Phe4
with the four isomers of â-methylphenylalanine, fol-
lowed by NMR and molecular modeling studies of these
derivatives, led to the conclusion that the gauche (-)
conformation was required for interaction of the Phe4
side chain with the δ-opioid receptor.16 Further sub-
stitution of Tyr1 in DPDPE with the four isomers of
â-methyl-2′,6′-dimethyltyrosine (TMT), along with de-
tailed computational and NMR studies, revealed that
the tyrosine derivative in the most potent and selective
analogue, [(2S,3R)-TMT1]DPDPE, preferred a trans
side-chain conformation.10 Molecular modeling studies
of [(2S,3R)-TMT1]DPDPE and two potent non-peptide
δ-opioid receptor agonists, such as SIOM6b and
2-m et h yl-4a R-(3-h ydr oxyph en yl)-1,2,3,4,4a ,5,12,-
12aR-octahydroquinolino[2,3,3-g]isoquinoline (TAN-67),7
resulted in a 3D pharmacophore model for this receptor.
From this model a characteristic distance of 7.0 ( 1.5
Å between the phenol and phenyl rings was observed.11
Computer-assisted template search indicated that two
cis 1,4-disubstituted benzyl-like aromatic rings can be
Resu lts a n d Discu ssion
Design . Detailed structure-activity relationship
studies on endogenous δ-opioid-selective linear peptide
ligands, enkephalins H-Tyr-Gly-Gly-Phe-Met(Leu)-OH
(Figure 1), have demonstrated that the phenol ring and
the amino group in the tyrosine residue and the phenyl
ring in the phenylalanine residue are the essential
pharmacophores for peptide ligands to recognize δ-opi-
oid receptors and produce physiological effects.4 The
substitution of both Gly2 and Met5 (or Leu5) residues
in enkephalin with D-Pen (â,â-dimethylcysteine), to form
a disulfide bridge, produced the potent agonist c[D-
Pen2,D-Pen5]enkephalin (DPDPE) which shows nano-
molar binding affinity at the δ-opioid receptor and over
350-fold selectivity for the δ- over µ-opioid receptor.12
In addition, DPDPE has no inhibitory effect on gut
motility, is stable to proteolytic degradation in blood,
brain, and gut, and is also able to penetrate through

Non-Peptide Agonists of the δ-Opioid Receptor
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4769
F igu r e 2. Design of the first generation of non-peptide mimetic ligands from the δ-opioid-selective peptide lead [(2S,3R)-TMT1]-
DPDPE.
Sch em e 1. Synthetic Route for the Preparation of Non-Peptide Mimetics 6a -f
presented at such distances on a cyclohexane-like scaf-
fold and that the hydroxy group of the phenol ring
should be in a meta position to overlap with the
corresponding pharmacophore in the bioactive confor-
mation of [(2S,3R)-TMT1]DPDPE.
In this first series of non-peptide ligands, a primary
amino group was omitted to simplify the synthetic work.
Furthermore, the nitrogen in the piperazine ring has
not been ideally positioned in 3D space to mimic the
terminal amino group in the peptide lead analogue
(Figure 2). Thus, we postulated that with three ap-
propriately oriented binding pharmacophore groups,
high binding affinity could still be expected according
to the theory proposed by Farmer.19 Therefore, the first
generation of peptidomimetic compounds was intention-
ally designed to explore the functional roles of the two
aromatic rings and the hydrophobic group R in the
ligand-receptor interaction.
Ch em istr y. The synthetic route for the preparation
of the non-peptide ligands 6a -f is shown in Scheme 1.
The hydroxyl group of 1 was protected as the methoxy-
ethoxymethyl ether (MEM).20 Except for alcohol 3a (R
) H) which was prepared through direct reduction of
aldehyde 2 with sodium borohydride, all other alcohols
3b-f were synthesized by reaction of the aldehyde 2
Additionally, an amino group may be introduced on
the cyclohexane scaffold to overlap the corresponding
amino group of the peptide ligand as shown in Figure
2.11 It also has been shown that the increased hydro-
phobicity of the tyrosine derivatives and the â-dimethyl
groups in both D-Pen residues may contribute to the
enhanced selectivity and binding affinity of the peptide
ligand.11,17,18 Thus, we introduced additional substitu-
ents R, of variable size and hydrophobicity, attached to
the benzyl carbon of the phenol side-chain group as
shown in Figure 2, to test how this would affect the
selectivity and binding affinity of these designed pep-
tidomimetics. To facilitate the synthetic process, we
used 1,4-piperazine as a 6-membered ring scaffold for
the first generation of peptidomimetics.

4770 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Ta ble 1. Binding Affinity of δ-Selective Opioid Peptide Ligands
Liao et al.
binding data, IC50 (nM) ( SEM
compound
[3H]DAMGO (µ)
[3H]p-Cl-DPDPE (δ)
selectivity (µ/δ)
860
1.3
[(2S,3R)TMT1]DPDPE3
6a (R ) H)
4300 ( 820
8100 ( 790
780 ( 72
2100 ( 600
500 ( 52
5.0 ( 0.1
6400 ( 3200
610 ( 310
420 ( 38
34 ( 17
6b (R ) Me)
1.3
5.0
15
6c (R ) i-Bu)
6d (R ) Ph)
6e (R ) C6H5Ph)
6f (R ) t-BuPh) (SL-3111)
(+)-SL-3111
∼27000
31 ( 4.9
∼870
2020
260
17000 ( 3000
11000 ( 1950
7700 ( 350
8.4 ( 1.6
42 ( 3.0
(-)-SL-3111
4.1 ( 1.0
1900
with the corresponding Grignard reagents. Alcohols
3a -f were converted into the chlorides 4 under neutral
conditions in the presence of triphenylphosphine.21
Compounds 4a -f were reacted with 1-benzylpiperazine
in the presence of potassium carbonate under overnight
reflux to produce compounds 5a -f. The hydroxyl
protecting group (MEM) was then removed with 2 N
HCl in a 1:1 methanol/dioxane mixture, by overnight
reaction, to give the desired final products 6a -f as white
solid hydrochloride salts.
ligands, just as the â-dimethyl groups of the D-Pen
residues in DPDPE are essential for selective binding
to the δ-receptor. The two optically pure enantiomers
of SL-3111 were synthesized and tested in the radio-
labeled ligand binding assays (Table 1). The enantiomer
(-)-SL-3111 has a very potent binding affinity for the
δ-opioid receptor (IC50 ) 4.1 nM), which is about 10-
fold more potent than (+)-SL-3111, with essentially the
same selectivity for the δ-opioid receptor (1900-fold) as
the racemic SL-3111 (2000-fold), due to the increased
binding affinity for the µ-opioid receptor. It is obvious
that the chirality at this center has an important effect
on the opioid ligand-receptor interactions.
Bin d in g to th e Wild -Typ e (Wt) a n d th e Mu ta ted
(W284L) Hu m a n δ-Op ioid Recep tor s. To determine
if the designed non-peptide ligand SL-3111 indeed
mimics the binding characteristics of the peptide ligand
toward the δ-receptor, the binding profiles of SL-3111,
[p-ClPhe4]DPDPE, and the non-peptide ligand SNC-805
were examined with the cloned human δ-opioid receptor
(wild-type, Wt) and the mutated human δ-opioid recep-
tor W248L.22 As shown in Figure 3, the binding curves
of [p-ClPhe4]DPDPE with Wt and W248L receptors were
essentially identical, showing IC50 values of 1.53 and
1.55 nM for Wt and W248L receptors, respectively
(Table 2). The binding curves of SL-3111 with the Wt
and W248L receptors had only a slight difference,
showing IC50 values of 5.5 and 16.8 nM with Wt and
W248L, respectively; the 3-fold difference was not found
to be significant. However, the binding curves of SNC-
80 with these receptors showed important differences,
with IC50 values of 2.8 nM at the Wt δ-opioid receptor
and 49.1 nM at the W248L δ-opioid receptor. These
results are consistent with the recent pharmacological
studies on the binding of δ-opioid-selective peptide and
non-peptide ligands to the wild-type and W284L mutant
human δ-opioid receptors.22
Str u ctu r e-Activity Rela tion sh ip s. All the syn-
thesized non-peptide ligands were evaluated using
standard binding assays against radiolabeled δ- and
µ-receptor-selective opioid ligands. The binding assay
results are given in Table 1. As expected, different
trends related to the size and hydrophobicity of the R
groups present in these non-peptide mimetics were
observed for the µ- and δ-opioid receptor binding affini-
ties. In the case of the µ-opioid receptor an especially
weak binding affinity was observed for compounds
without a substituent 6a (R ) H) or with a bulky
substituent (R ) i-Bu, p-C6H5Ph, t-BuPh, 6c,e,f). On
the other hand, a substituent of intermediate size
(methyl group 6b or a planar phenyl group 6d ) produced
compounds with higher binding affinities. These results
suggest that a hydrophobic but less bulky R group may
present a better fit to the lipophilic pocket of the µ-opioid
receptor. For the δ-opioid receptor, it was observed that
an increase in binding affinity and selectivity was
proportional to the hydrophobicity and size of the
substituent R in compounds 6a -f (Table 1). However,
for compound 6d (R ) phenyl) the substitution with a
more bulky and hydrophobic R group (R ) C6H5Ph) to
produce 6e did not improve the binding affinity of the
non-peptide ligand for the δ-opioid receptor but signifi-
cantly reduced the binding affinity for the µ-opioid
receptor as described above. Substitution of the p-
phenyl group in 6e with a bulkier p-tert-butyl group led
to the analogue 6f (SL-3111), which showed a 4-fold
increase on binding affinity for the δ-opioid receptor
(IC50 ) 8.4 nM) relative to 6e and is comparable to
[(2S,3R)-TMT1]DPDPE.10 In addition, 6f showed very
weak affinity for the µ-receptor. This makes 1-(4-tert-
butyl-3′-hydroxybenzhydryl)-4-benzylpiperazine (6f, SL-
3111) one of the most selective non-peptide mimetic
ligands reported so far for binding to the δ-opioid
receptor in preference to the µ-opioid receptor with a
2000-fold µ/δ selectivity. Indeed, this exceeds the
selectivity of the peptide lead [(2S,3R)-TMT1]DPDPE.
Apparently, as was postulated in the design, a bulky
hydrophobic R group in SL-3111 may mimic an impor-
tant lipophilic binding site in the δ-opioid peptide
The W284L mutation led to a receptor with pro-
foundly reduced affinity to the non-peptide opioid ligand
SNC-121, but one which did not alter binding affinities
of typical δ-opioid peptide agonists such as [p-ClPhe4]-
DPDPE and deltorphin I. The high degree of ligand
specificity shown by the W284L mutation supports a
conclusion that the amino acid in this position of the
receptor is involved in a direct interaction with the SNC
series of compounds that does not occur for peptide and
peptidomimetic δ-opioid-selective ligands. Although
chemically related to SNC-80, SL-3111 has a binding
profile more closely related to that of the classical
peptide [p-ClPhe4]DPDPE than SNC-80 discovered from
screening methods.

Non-Peptide Agonists of the δ-Opioid Receptor
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4771
F igu r e 3. Inhibition of specific [3H]p-Cl-DPDPE binding by selective opioid agonists: p-Cl-DPDPE, SL-3111, and SNC-80. The
membrane cells were transfected with the cDNAs encoding the Wt hDOR (9) and W248L mutant hMOR (b). The specific [3H]-
p-Cl-DPDPE binding to membrane preparations was inhibited by increasing concentration of p-Cl-DPDPE, SL-3111, and SNC-
80.
Ta ble 2. Effect of the Mutation of Cloned Human δ-Opioid
Receptor on the Binding Affinity (IC50, nM) of δ-Selective
Opioid Ligandsa
In Vitr o Bioa ssa ys. Compounds 6d ,f (SL-3111)
show nanomolar binding affinities for the δ-opioid
receptor in the rat brain and consequently were evalu-
ated in the isolated mouse vas deferens (MVD) and
guinea pig ileum (GPI) bioassays (Table 4). Compared
to the binding affinity, 6d shows a 5- and 2.5-fold
decrease of potency in the MVD (δ) and GPI (µ) assays,
respectively. On the other hand, compound 6f (SL-3111)
shows a 10-fold decrease of potency in the MVD (δ)
assay and a 2-fold decrease in the GPI (µ) assay, still
maintaining a high (460-fold) selectivity. The relatively
low biological potency may be related to the possible
improper position of the nitrogen atoms in these non-
peptide analogues, which are unable to interact with
the same site at the δ-receptor as the terminal amino
group of the peptide lead does. In these assays, SL-
3111 was used as the racemic mixture. Therefore, we
sought to evaluate the biological activities of the enan-
tiomers. However, when the optically pure isomers
were tested, a 2.4- and 4.2-fold decrease in potency in
MVD (δ) assay was observed for the (+)- and (-)-SL-
3111 enantiomer, respectively (Table 4). As shown in
Table 4, the peptide lead [(2S,3R)-TMT1]DPDPE pos-
sesses a high potency in the δ-receptor bioassay and
antagonism for the µ-receptor,10 which makes it highly
selective for the δ-opioid receptor. The difference in the
binding affinity and biological potency between [(2S,3R)-
TMT1]DPDPE and the non-peptide mimetic SL-3111
raises interesting questions about the relationships
between binding affinity and biological efficacy,23 on the
chemical-structural basis for binding affinity, and if
additional features are required for the expected bio-
logical response. Understanding these fundamental
issues are necessary for successful rational drug design.
Thus, we believe that SL-3111 and the related pepti-
domimetic ligands presented in this paper show poten-
tial use as tools to explore the stereochemical require-
ments for δ-opioid-selective non-peptide ligands, providing
new leads to develop novel δ-selective non-peptide
mimetics.
ligand
Wt
W284L
W284L/Wt
p-Cl-DPDPE
SL-3111
SNC-80
1.53 ( 0.26
5.51 ( 0.94
2.85 ( 0.26
1.55 ( 0.11
16.8 ( 6.13
49.1 ( 4.33
1.0
3.1
17.2
a
Wt, cloned human δ-opioid receptor; W284L, mutated human
δ-opioid receptor in which the Trp residue at position 284 was
mutated to Leu.
Effects of th e Meth yla tion of th e P h en ol Gr ou p
on th e Bin d in g Affin ity of δ-Op ioid -Selective P ep -
tid e a n d Non -P ep tid e Liga n d s. Although the hy-
droxy functional group of the tyrosine residue in en-
kephalins has been identified to be a crucial pharmaco-
phore for binding to the δ-opioid receptor, its importance
in DPDPE itself has not been examined. Since methy-
lation of (+)-4-[((RR)-R(2S,5R)-4-allyl-2,5-dimethyl-1-
piperazinyl)-3-hydroxybenzyl]-N,N-diethylbenzamide [(+)-
BW373U86] produces an extremely selective δ-opioid
analogue (SNC-80), with over 2000-fold selectivity for
the δ-opioid receptor while maintaining nearly the same
binding affinity as (+)-BW373U86,5 the hydroxyl group
has not been thought to be an essential pharmacophore
for these δ-opioid-selective ligands. Thus, to identify
the importance of the hydroxyl groups in traditional
δ-opioid-selective peptide ligands, a new peptide ana-
logue ([p-MeOTyr1]DPDPE) was synthesized. The bio-
assay results of this peptide analogue were compared
with those from the O-methylated analogue of SL-3111
(6g). As shown in Table 3, methylation of the hydroxyl
group of the Tyr1 residue has a considerable effect on
the binding affinity of DPDPE, causing a 150-fold drop
in the binding affinity of [p-MeOTyr1]DPDPE for the
δ-opioid receptor. The same effect was observed for 6g.
These results clearly demonstrate that the hydroxyl
group in Tyr1 residue of the potent δ-opioid-selective
peptide DPDPE and in the phenol group of SL-3111 are
crucial binding pharmacophores, which probably inter-
act with the same binding sites of the δ-opioid receptor.
However, this is not the case for the SNC series of non-
peptide ligands for which methylation did not affect the
binding affinity to the δ-opioid receptor as discussed
above. These results suggest once again that the
binding profile of SL-3111 is more closely related to that
of the peptide ligand DPDPE than the SNC series of
non-peptide ligands, consistent with our design pro-
posal.
Con clu sion s
A new δ-opioid-selective non-peptide agonist lead (SL-
3111) has been discovered based on the structure-
activity relationships of a δ-opioid peptide agonist
ligand. This non-peptide ligand showed a binding
affinity of 8 nM and over 2000-fold selectivity for the
δ-opioid receptor over the µ-opioid receptor. The chiral

4772 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Liao et al.
Ta ble 3. Effect of Methoxylation of the Phenol Groups on the Binding Affinities (IC50, nM) of δ-Selective Opioid Ligands
ligand
[3H]DAMGO (µ)
[3H]p-Cl-DPDPE (δ)
selectivity (µ/δ)
DPDPEa
[MeOTyr1]DPDPE
609 ( 70
11000 ( 2700
17000 ( 3000
1.6 ( 0.2
230 ( 24
8.4 ( 1.6
380
48
2020
>8000
1800
>4
9.7 ( 0.37
0.31 ( 0.02
31
2500 ( 200
1.06 ( 0.14
2300
a
b
See refs 16, 17, and 27. See ref 5.
Ta ble 4. Biological Potencies of the Non-Peptide Mimetics
bioassay data, EC50 (nM) ( SEM
MVD (δ)
compound
GPI (µ)
0% at 60 µM
selectivity (µ/δ)
>33000
[(2S,3R)TMT1]DPDPEc
1.8 ( 0.3
(antagonist, IC50 ) 5 µM)
1250 ( 240
6d
174 ( 15.5
7.1
CTAPa insensitive, ICI-174,864b sensitive
SL-3111 (6f)
39000 ( 2600
85 ( 10
460
CTAPa insensitive, ICI-174,864b sensitive
(+)-SL-3111
(-)-SL-3111
3000 ( 1500
7600 ( 6500
210 ( 50
360 ( 57
18.5
21.0
a
b
A peptide µ-opioid receptor antagonist. A δ-selective enkephalin-derived peptide antagonist. 6d was shifted 1.3-fold by 1 µM CTAP
and 6.0-fold by 1 µM ICI-174,864. The dose-response curve of SL-3111 (6f) was shifted to the right 2.6-fold by 1 µM CTAP and 21.8-fold
by 1 µM ICI-174,864. c See ref 10.
HPLC resolution of the racemic SL-3111 provided the
(+)- and (-)-enantiomers which were 5-fold less potent
and 2-fold more potent, respectively, in our binding
assay. It was observed that the hydroxyl group of the
phenol pharmacophore in SL-3111 was essential to
maintain high potency for the δ-opioid receptor, as was
also found for the potent δ-opioid-selective peptide
ligand DPDPE. Further binding studies of SL-3111 and
the δ-opioid-selective peptide ligand [p-ClPhe4]DPDPE
on the cloned wild-type human and mutated δ-opioid
receptors suggested that both present similar binding
profiles which differ from those of another δ-opioid-
selective non-peptide ligand (SNC-80). This makes SL-
3111 a true peptidomimetic which provides a new lead
structure for further design of new generations of
δ-opioid non-peptide ligands.
Exp er im en ta l Section
P r ep a r a tion of Non -P ep tid e Liga n d s. All reagents,
unless otherwise noted, were purchased from Aldrich Chemical

Non-Peptide Agonists of the δ-Opioid Receptor
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4773
Co. and were used without further purification. All melting
points were taken on a Thomas-Hoover capillary melting point
apparatus and are uncorrected. NMR spectra were recorded
with a Bruker AM 250 spectrometer, using tetramethylsilane
(TMS) or D2O (4.66 ppm downfield from TMS) as an internal
standard. Optical rotations were measured on an Autopol III
polarimeter using a 1.0-dm cell. Flash column chromatogra-
phy was performed using E. Merck silica gel (230 mesh).
Purification of the peptide was achieved by preparative
reverse-phase high-performance liquid chromatography (HPLC)
on a Perkin-Elmer instrument model 410-Bio, C18-bonded
silica column (VYDAC 218TP1010, 12 × 275 mm), eluting with
a linear gradient of acetonitrile in 0.1% aqueous TFA (10-
90%) over 30 min at a flow rate of 5 mL/min. High-resolution
mass spectra (HRMS) were obtained at the University of
Arizona Chemistry Department Mass Spectroscopy Facility,
with a J EOL mass spectrometer model HX-110. All analytical
(C, H, N) and spectroscopic (1H NMR, 13C NMR, MS) data are
in agreement with the proposed structures.
tectin g Gr ou p (Com p ou n d s 6a -g). HCl 2 N (2 mL/0.1 g)
was added into a solution of 5a dissolved in a 1:1 MeOH/
dioxane mixture. The solution was stirred at room tempera-
ture for 24 h, and the volatiles were removed in vacuo at room
temperature. To the residue was added cold ether to precipi-
tate the product. If no precipitate was formed, the residue
was redissolved in ethyl acetate and hexane was added to
precipitate the product. The white precipitate was filtered off,
washed with ether, and dried in vacuo: hydrochloride salt,
white solid (95%); 1H NMR (CD3OD) δ 7.32-6.78 (m, 9H), 4.27
(s, 2H), 4.20 (s, 2H), 3.43 (bs, 8H). 13C NMR (D2O) δ 157.2,
132.7, 132.0, 131.7, 130.5, 130.1, 128.5, 124.1, 118.9, 118.5,
61.5, 61.2, 49.1, 36.1; FAB-HRMS calcd for C18H23N2O [M +
H]+ ) 283.1810, found 283.1808. Anal. Calcd for C18H22N2O‚
2HCl: C, 61.06; H, 6.84; N, 7.91. Found: C, 60.01; H, 7.03; N,
7.42.
Syn th esis of 1-(3-Hydr oxy-r-m eth ylben zyl)-4-ben zylpip-
er a zin e (6b) fr om 2. P r ep a r a tion of 1-[3-(2-Meth oxy-
eth oxym eth yl)p h en yl]eth a n ol (3b). Into a stirred solution
of methylmagnesium bromide (3.0 M, 4.5 mL, 5 equiv) in 10
mL of THF at 0 °C was added dropwise a solution of 2 in 5
mL of THF. The resulting gray suspension was allowed to
warm to room temperature and monitored by TLC. Then, the
reaction was quenched by addition of 0.5 N HCl (30 mL). The
product was extracted with ether (3 × 40 mL), and the
combined organic fractions were washed with brine (60 mL),
water (60 mL) and and dried over anhydrous MgSO4. Evapo-
ration of the solution gave the product which was further
purified by flash chromatography (EtOAc/hexanes, 2:8): color-
less oil (95%); 1H NMR (CDCl3) δ 7.30-6.998 (m, 4H), 5.27 (s,
2H), 4.87 (q, 1H, J ) 6.5 Hz), 3.85-3.81 (m, 2H), 3.58-3.55
(m, 2H), 3.38 (s, 3H), 1.97 (bs, 1H), 1.48 (d, 3H, J ) 6.8 Hz).
P r ep a r a tion of 1-Ch lor o-1-[3-(2-m eth oxyeth oxym eth -
yl)p h en yl]eth a n e (4b). Following the same procedure out-
lined for 4a (vide supra) but starting with 3b led to 4b as a
colorless oil (60%): 1H NMR (CDCl3) δ 7.16-6.87 (m, 4H), 5.17
(s, 2H), 4.93 (q, 1H), 3.74-3.70 (m, 2H), 3.47-3.43 (m, 2H),
3.27 (s, 3H), 1.72 (d, 3H, J ) 6.8 Hz).
3-(2-Meth oxyeth oxym eth yl)ben za ld eh yd e (2). Into a
stirred solution of 3-hydroxybenzaldehyde (3.7 g, 30 mmol) in
60 mL of DCM, was added DIEA (7.6 mL, 43.7 mmol). To the
resulting brown solution was added MEMCl (5.0 mL, 43.7
mmol) in 17 mL of DCM dropwise. A white fog was observed
in the addition process. The resulting solution was stirred at
room temperature for 3 h and then quenched by addition of
60 mL of 0.5 N HCl. The organic phase was separated and
the aqueous phase extracted with DCM (3 × 60 mL). The
combined organic phases were washed with 5% Na2CO3 (3 ×
60 mL) and water (3 × 60 mL) and then dried over anhydrous
MgSO4. Evaporation of the dried solution yielded 5.4 g of an
orange oil (86%): 1H NMR (CDCl3) δ 9.90 (s, 1H), 7.57-7.45
(m, 4H), 5.33 (s, 2H), 3.86-3.82 (m, 2H), 3.58-3.54 (m, 2H),
3.38 (s, 3H); 13C NMR (CDCl3) δ 191.8, 157.7, 137.7, 130.0,
123.5, 122.6 93.3, 71.4, 67.7, 58.9; HRMS-FAB calcd for C11
15O4 [M + H]+ ) 211.0970, found 211.0973.
3-(2-Meth oxyeth oxym eth yl)ben zylic Alcoh ol (3a ). Into
-
H
a stirred solution of 2 (1 g, 8.2 mmol) in 30 mL of ethanol was
added sodium borohydride powder until the bubbling stopped.
Then the reaction was quenched with 15 mL of 2 N HCl. The
product was extracted with ether (3 × 30 mL). The combined
organic extracts were washed with NaHCO3 (40 mL) and brine
(40 mL) and dried over anhydrous MgSO4. Evaporation
yielded a light-yellow oil, which was purified by flash chro-
matography (EtOAc/hexane, 3:7), to give a colorless oil (95%):
1H NMR (CDCl3) δ 7.30-6.95 (m, 4H), 5.27 (s, 2H), 4.65 (s,
2H), 3.84-3.80 (m, 2H), 3.57-3.54 (m, 2H), 3.37 (s, 3H), 1.94
(bs, 1H); 13C NMR (CDCl3) δ 157.4, 142.8, 129.6, 120.3, 115.4
114.7, 93.3, 71.5, 67.6, 65.1, 59.0.
Syn th esis of 1-(3-Hyd r oxyben zyl)-4-ben zylp ip er a zin e
(6a ) fr om 3a . P r ep a r a tion of 3-(2-Meth oxyeth oxym eth -
yl)ben zyl Ch lor id e (4a ). A mixture of 3a (2.38 mmol) and
triphenylphosphine (3.43 mmol) in 6.0 mL of carbon tetra-
chloride was refluxed under Ar for 3 h. A white precipitate
was formed. After the mixture cooled to room temperature,
10 mL of anhydrous ether was added to the mixture. The solid
was filtered off and washed with ether (3 × 10 mL). The
combined filtrate was evaporated to dryness and purified by
flash chromatography (AtOAc/hexane, 3:7): colorless oil (20%);
1H NMR (CDCl3) δ 7.30-6.99 (m, 4H), 5.27 (s, 2H), 4.55 (s,
2H), 3.85-3.81 (m, 2H), 3.58-3.54 (m, 2H), 3.38 (s, 3H).
P r ep a r a tion of 1-[3-(2-Meth oxyeth oxym eth yl)-r-m e-
th ylben zyl]-4-ben zylp ip er a zin e (5b). Following the same
method as used to prepare 5a but starting from 4b led to 5b
as a colorless oil (62%): 1H NMR (CDCl3) δ 7.30-6.90 (m, 9H),
5.26 (s, 2H), 3.85-3.81 (m, 2H), 3.49 (s, 2H), 3.37 (s, 3H), 3.35
(q, 1H, J ) 6.7 Hz), 2.44 (bs, 8H), 1.34 (d, 3H, J ) 6.7 Hz).
Syn th esis of 1-(3-Hydr oxy-r-m eth ylben zyl)-4-ben zylpip-
er a zin e (6b). Following the same method as used to prepare
6a but starting with 5b led to 6b as the hydrochloride salt:
white solid (95%); 1H NMR (CD3OD) δ 7.34-6.81 (m, 9H), 4.34
(q, 1H J ) 6.9 Hz), 4.28 (s, 2H), 3.45 (s, 4H), 3.30 (bs, 4H),
1.58 (d, 3H, J ) 6.8 Hz); 13C NMR (CDCl3) δ 156.0, 142.7,
136.5, 129.7, 129.0, 128.3, 127.3, 119.6, 115.0, 114.6, 64.3, 63.0,
53.0, 49.3, 19.0; FAB-HRMS calcd for C19H25N2O [M + H]+
)
297.1967, found 297.1972. Anal. Calcd for C19H24N2O: C,
76.98; H, 8.19; N, 9.46. Found: C, 77.07; H, 8.00; N, 9.20.
Syn t h esis of 1-(3-H yd r oxy-R-isob u t ylb en zyl)-4-b en z-
ylp ip er a zin e (6c). P r ep a r a tion of 1-[3-(2-Meth oxy-
eth oxym eth yl)p h en yl]-3-m eth ylbu ta n ol (3c). Following
the same general procedure outlined for 3b but using iso-
butylmagnesium bromide led to 3c as a colorless oil (45%):
1H NMR (CDCl3) δ 7.27-6.93 (m, 4H), 5.27 (s, 2H), 4.69 (m,
1H), 3.83-3.79 (m, 2H), 3.56-3.52 (m, 2H), 3.37 (s, 3H), 1.94-
1.60 (m, 2H), 1.49-1.45 (m, 1H), 0.94 (d, 6H, J ) 6.3 Hz).
P r ep a r a tion of 1-Ch lor o-1-[3-(2-m eth oxyeth oxym eth -
yl)ph en yl]-3-m eth ylbu tan e (4c). Following the same method
as used to prepare 4a (vide supra) but starting with 3c led to
4c as a colorless oil (80%): 1H NMR (CDCl3) δ 7.29-6.97 (m,
4H), 5.27 (s, 2H), 4.88 (dd, 1H, J ) 6.0 Hz), 3.85-3.81 (m,
2H), 3.57-3.53 (m, 2H), 3.37 (s, 3H), 2.12-1.95 (m, 1H), 1.84-
1.76 (m, 2H), 0.95 (dd, 6H, J ) 6.4 Hz).
P r ep a r a tion of 1-[3-(2-Meth oxyeth oxym eth yl)ben zyl]-
4-ben zylp ip er a zin e (5a ). A mixture of 1-benzylpiperazine
(0.8 g, 4.8 mmol), potassium carbonate (0.4 g, 2.9 mmol), and
4a (1.3 mmol) in 6.0 mL of acetonitrile was refluxed under Ar
for 2 h. After cooling to room temperature, the solid was
filtered out and washed with acetonitrile (3 × 10 mL). The
filtrates were combined and concentrated to yield a yellow oil,
which was purified by flash chromatography (EtOAc/hexane/
1
Et3N, 30:70:1): colorless oil (80%); H NMR (CDCl3) δ 7.31-
P r ep a r a t ion of 1-[3-(2-Met h oxyet h oxym et h yl)-r-iso-
bu tylben zyl]-4-ben zylp ip er a zin e (5c). Following the same
procedure as described for 5a but starting with 4c led to 5c
as a colorless oil (22%): 1H NMR (CDCl3) δ 7.27-6.81 (m, 9H),
5.26 (s, 2H), 3.84-3.81 (m, 2H), 3.57-3.55 (m, 2H), 3.45 (s,
6.93 (m, 9H), 5.26 (s, 2H), 3.84-3.80 (m, 2H), 3.57-3.53 (m,
2H), 3.50 (s, 2H), 3.48 (s, 2H), 3.37 (s, 3H), 2.47 (bs, 8H).
P r ep a r a tion of 1-(3-Hyd r oxyben zyl)-4-ben zylp ip er a -
zin e (6a ). Gen er a l Meth od for Clea va ge of MEM P r o-

4774 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Liao et al.
2H), 3.40-3.36 (m, 4H), 2.43 (bs, 8H), 1.68 (dd, 2H, J ) 6.7
Hz), 1.35 (m, 1H), 0.84 (t, 6H, J ) 6.3 Hz).
ben zylp ip er a zin e (6f) (SL-3111) fr om 2. P r ep a r a tion of
4-ter t-Bu tyl-3-(2-m eth oxyeth oxym eth yl)ben zh yd r ol (3f).
Following the same general procedure outlined for 3b but
using 4-tert-butylphenylmagnesium bromide led to 3f as a
colorless oil (80%): 1H NMR (CDCl3) δ 7.36-6.76 (m, 8H), 5.77
(d, 1H), 3.78 (s, 3H), 2.26 (bs, 1H), 1.29 (s, 9H).
P r ep a r a tion of 4-ter t-Bu tyl-3-(2-m eth oxyeth oxym eth -
yl)ben zh yd r yl Ch lor id e (4f). Following the same method
as used to prepare 4a (vide supra) but starting with 3f led to
4f as a colorless oil (70%): 1H NMR (CDCl3) δ 7.34-6.80 (m,
8H), 6.08 (s, 1H), 3.75 (s, 3H), 1.29 (s, 9H).
P r ep a r a t ion of 1-[4-ter t-Bu t yl-3′-(2-m et h oxyet h oxy-
m eth yl)ben zh yd r yl]-4-ben zylp ip er a zin e (5f). Following
the same procedure as described for 5a but starting with 4f
led to 5f as a colorless oil (70%): 1H NMR (CDCl3) δ 7.34-
7.68 (m, 13H), 4.13 (s, 1H), 3.77 (s, 3H), 3.50 (s, 2H), 2.45 (bs,
8H), 1.28 (s, 9H).
Syn th esis of 1-(4-ter t-Bu tyl-3′-h yd r oxyben zh yd r yl)-4-
ben zylpiper azin e (6f) (SL-3111). Following the same method
as used to prepare 6a but starting with 5f led to 6f as the
hydrochloride salt: white solid (83%); 1H NMR (CD3OD) δ
7.73-7.21 (m, 13H), 4.49 (s, 1H), 3.73 (s, 2H), 3.63 (bs, 4H),
3.55 (s, 4H), 1.20 (s, 9H); 13C NMR (CDCl3) δ 157.9, 152.9,
131.2, 130.5, 129.5, 129.4, 128.0, 127.2, 126.7, 119.8, 117.3,
Syn t h esis
of
1-(3-H yd r oxy-r-isob u t ylb en zyl)-4-
ben zylp ip er a zin e (6c). Following the same method as used
to prepare 6a but starting with 5c led to 6c as the hydrochlo-
ride salt: white solid (95%); 1H NMR (CD3OD) δ 7.35-6.80
(m, 9H), 4.24 (m, 3H), 3.40 (s, 4H), 3.20 (bs, 4H), 2.00 (m, 1H),
1.03 (m, 1H), 0.69 (d, 3H, J ) 9.8 Hz), 0.66 (d, 3H, J ) 9.6
Hz); FAB-HRMS calcd for C22H31N2O [M + H]+ ) 339.2436,
found 339.2425. Anal. Calcd for C22H30N2O‚2HCl: C, 64.45;
H, 7.80; N, 6.83. Found: C, 68.83; H, 8.80; N, 6.50.
Syn t h esis of 1-(3-Hyd r oxyb en zh yd r yl)-4-b en zylp ip -
er a zin e (6d ) fr om 2. Preparation of 3-(2-Methoxyethoxy-
methyl)benzhydrol (3d). Following the same general procedure
outlined for 3b but using phenylmagnesium bromide led to
3d as a colorless oil (95%): 1H NMR (CDCl3) δ 7.35-6.85 (m,
9H), 5.75 (s, 1H), 5.24 (s, 2H), 3.78-3.72 (m, 2H), 3.41-3.48
(m, 2H), 3.35 (s, 3H), 2.30 (bs, 1H).
P r ep a r a tion of 3-(2-Meth oxyeth oxym eth yl)ben zh yd r -
yl Ch lor id e (4d ). Following the same method as used to
prepare 4a (vide supra) but starting with 3d led to 4d as a
colorless oil (65%): 1H NMR (CDCl3) δ 7.43-7.06 (m, 9H), 6.08
(s, 1H), 5.25 (s, 2H), 3.83-3.78 (m, 2H), 3.54-3.52 (m 2H),
3.36 (s, 3H).
P r ep a r a tion of 1-[3-(2-Meth oxyeth oxym eth yl)ben zh y-
d r yl]-4-ben zylp ip er a zin e (5d ). Following the same proce-
dure as described for 5a but starting with 4d led to 5d as a
colorless oil (77%): 1H NMR (CDCl3) δ 7.40-6.84 (m, 14H),
5.23 (s, 2H), 4.18 (s, 1H), 3.80-3.77 (m, 2H), 3.54-3.50 (m,
4H), 3.35 (s, 3H), 2.44 (bs, 8H).
Syn t h esis of 1-(3-Hyd r oxyb en zh yd r yl)-4-b en zylp ip -
er a zin e (6d ). Following the same method as used to prepare
6a but starting with 5d led to 6d as the hydrochloride salt:
white solid (95%); 1H NMR (CD3OD) δ 7.37-6.58 (m, 14H),
4.17 (s, 1H), 3.53 (s, 2H), 2.48 (bs, 8H), 2.00 (m, 1H), 1.03 (m,
1H), 0.69 (d, 3H, J ) 9.8 Hz), 0.66 (d, 3H, J ) 9.6 Hz); 13C
NMR (DMSO-d6) δ 157.9, 137.2, 136.1, 131.5, 130.4, 129.6,
129.2, 129.0, 128.8, 128.3, 118.5, 115.8, 115.3, 57.3, 45.4, 32.9;
FAB-HRMS calcd for C24H27N2O [M + H]+ ) 359.2123, found
359.2128. Anal. Calcd for C24H26N2O‚2HCl: C, 67.03; H, 6.51;
N, 6.51. Found: C, 66.15; H, 6.71; N, 6.30.
Syn t h esis of 1-(3-H yd r oxy-4′-p h en ylb en zh yd r yl)-4-
ben zylp ip er a zin e (6e) fr om 2. P r ep a r a t ion of 3-(2-
Meth oxyeth oxym eth yl)-4′-p h en ylben zh yd r ol (3e). Fol-
lowing the same general procedure outlined for 3b but using
4-phenylphenylmagnesium bromide led to 3e as a colorless oil
(85%): 1H NMR (CDCl3) δ 7.36-6.98 (m, 8H), 5.77 (d, 1H, J
) 3.5 Hz), 5.24 (s, 2H), 3.82-3.78 (m,2H), 3.55-3.51 (m, 2H),
3.35 (s, 3H), 2.29 (s, 1H, J ) 3.5 Hz), 1.29 (s, 9H).
114.5, 112.9, 48.4, 48.3, 34.6, 31.0; FAB-HRMS calcd for C28
H
for C28H34N2O‚2HCl: C, 69.06; H, 7.46; N, 5.75. Found: C,
66.40; H, 7.56; N, 5.31.
-
35N2O [M + H]+ ) 415.2749, found 415.2742. Anal. Calcd
Syn th esis of 1-(4-ter t-Bu tyl-3′-m eth oxyben zh yd r yl)-4-
ben zylp ip er a zin e (6g). Following the same procedure de-
scribed above for the preparation of analogues 6b-f but using
3-methoxybenzaldehyde instead of 2 led to 6g as the hydro-
chloride salt: white solid (85%); 1H NMR (CDCl3) δ 7.54-6.83
(m, 13H), 4.70 (s, 1H), 4.25 (s, 2H), 3.80 (s, 3H), 3.62 (m, 4H),
3.33 (bs, 4H); 13C NMR (CDCl3) δ 160.5, 152.5, 137.0, 132.2,
130.9, 130.5, 71.8, 62.08, 60.4, 55.3, 48.6, 34.6; FAB-MS calcd
for C29H36N2O [M + H]+ ) 429.2906, found 429.2902. Anal.
Calcd for C29H36N2O‚2CF3CO2H: C, 60.41; H, 5.85; N, 4.27.
Found: C, 60.22; H, 5.81; N, 4.08.
The two optically pure enantiomers of SL-3111 (6f) were
obtained by HPLC separation performed on a Shimadzu liquid
chromatograph model SCL-10A, using a chiral column Chiral-
pack-AD (DAICEL) (25 cm × 0.5 cm). An isocratic eluent
mixture of hexane/2-propanol (97:3) was used as mobile phase
with the detector set at 254 nm. The concentration of each
injection was 1.5 mg/mL to obtain an adequate resolution. (+)-
22
SL-3111: white solid; mp 109-110 °C; tR ) 7.2 min; [R]D
)
+11.5 (c 1, CHCl3). (-)-SL-3111: white solid; mp 109-110
22
°C; tR ) 8.1 min; [R]D ) -10.6 (c 1, CHCl3).
P r ep a r a tion of [p-MeOTyr 1]DP DP E. This peptide was
synthesized on a solid support using the NR-Boc strategy. NR-
Boc-D-Pen(S-p-MeBzl)-resin (1 g, 0.4 mmol/g) was used as
starting material. The attachment of the NR-Boc-D-Pen(S-p-
MeBzl) to the chloromethylated Merrifield resin (Advanced
Chemtech, KY) was performed by Gisin’s method.24 The
following protected amino acids were assembled on a solid
support in a stepwise fashion to the growing peptide chain:
NR-Boc-Phe, NR-Boc-Gly, NR-Boc-D-Pen(S-p-MeBzl), and NR-
Boc-Tyr(OMe)-OH (Bachem, CA). The cyclization of the linear
peptide was achieved by procedures reported in the literature25
and purified by HPLC as described previously: tR ) 16.5 min;
HRMS-FAB (m-NBA matrix) calcd for C31H42N5O7S2 [M + H]+
) 660.2526, found 660.2529. The purity of the synthesized
peptide was also analyzed by TLC in three different solvent
systems and visualized by staining with 0.2% ninhydrin in
ethanol and heating:26 (A) Rf ) 0.70 (n-BuOH/AcOH/H2O,
P r ep a r a tion of 3-(2-Meth oxyeth oxym eth yl)-4′-p h en yl-
ben zh yd r yl Ch lor id e (4e). Following the same method as
used to prepare 4a (vide supra) but starting with 3e led to 4e
as a colorless oil (34%): 1H NMR (CDCl3) δ 7.345-7.00 (m,
8H), 6.06 (s, 1H), 5.25 (s, 2H), 3.82-3.70 (m, 2H), 3.64-3.50
(m, 2H), 3.35 (s, 3H), 1.29 (s, 9H).
P r ep a r a t ion
of 1-[3-(2-Met h oxyet h oxym et h yl)-4′-
p h en ylben zh yd r yl]-4-ben zylp ip er a zin e (5e). Following
the same procedure as described for 5a but starting with 4e
led to 5e as a colorless oil (70%): 1H NMR (CDCl3) δ 7.28-
7.12 (m, 13H), 5.23 (s, 2H), 4.15 (s, 1H), 3.81-3.77 (m, 2H),
3.35-3.49 (m, 2H), 3.48 (s, 2H), 3.35 (s, 3H), 2.45 (bs, 8H),
1.25 (s, 9H).
Syn t h esis of 1-(3-H yd r oxy-4-p h en ylb en zh yd r yl)-4-
ben zylp ip er a zin e (6e). Following the same method as used
to prepare 6a but starting with 5e led to 6e as the hydrochlo-
ride salt: white solid (80%); 1H NMR (CD3OD) δ 7.76-7.13
(m, 18H), 4.17 (s, 1H), 5.43 (s, 1H), 4.36 (s, 2H), 3.60 (bs, 4H),
3.45 (s, 4H); 13C NMR (DMSO-d6) δ 158.0, 140.5, 139.2, 137.3,
135.1, 131.5, 130.5, 129.7, 129.0, 128.8, 127.8, 127.5, 126.7,
118.5, 115.9, 115.3, 38.2; FAB-HRMS calcd for C30H31N2O [M
4:1:5); (B) Rf
)
0.75 (n-BuOH/pyridine/AcOH/H2O,
3.75:2.5:0.75:3); (C) Rf ) 0.95 (AcOEt/pyridine/AcOH/H2O,
6:1.5:2.1:1.1).
Ra d iola beled Liga n d Bin d in g Assa ys. Membranes were
prepared according to the following procedure: adult male
Sprague-Dawley rats (200-300 g) obtained from Harlan
Sprague-Dawley, Inc. (Indianapolis, IN) were sacrificed and
their brains immediately removed and placed on ice. The
+ H]+ ) 435.2436, found 435.2427. Anal. Calcd for C30H30
-
N2O‚2HCl: C, 71.20; H, 6.38; N, 5.53. Found: C, 69.66; H,
6.57; N, 5.31.
Syn th esis of 1-(4-ter t-Bu tyl-3-h yd r oxyben zh yd r yl)-4-

Non-Peptide Agonists of the δ-Opioid Receptor
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24 4775
Peptide Metabolism. In Peptide-Based Drug Design: Controlling
Transport and Metabolism; Taylor, M. D., Amidon, G. L., Eds.;
American Chemical Society: Washington, DC, 1995; pp 387-
421. (e) Hruby, V. J . Prospects For Peptidomimetic Drug Today.
Drug Discovery Today 1997, 2, 165-167.
whole brain was homogenized in 20 volumes of 50 mM Tris-
HCl stock buffer (pH 7.4) with a glass-Teflon homogenizer.
The homogenate was centrifuged (48000g for 15 min), resus-
pended, and preincubated (25 °C for 30 min) to remove
endogenous opioids. The homogenate was centrifuged and
resuspended again (0.5% final concentration). The cloned
human δ-opioid receptor for study of the binding profiles of
SL-3111 (6f), [Phe(p-Cl)4]DPDPE, and SNC-80 was prepared
based on literature method.22 Binding affinities of the com-
pounds were measured against [3H][p-Cl-Phe4]DPDPE ([3H]-
Tyr-c[D-Pen-Gly-(p-Cl)Phe-D-Pen-OH]) (41.0 Ci/mmol)27 and
[3H]DAMGO ([3H]Tyr-D-Ala-Gly-N-methyl-Phe-Gly-ol) (48.9 Ci/
mmol)28 (New England Nuclear) by a rapid filtration technique.
A solution of 0.75 nM [3H][p-Cl-Phe4]DPDPE or 1.0 nM [3H]-
DAMGO in a total volume of 1 mL of 50 mM Tris-HCl buffer
(pH 7.4) containing bovine serum albumin (1.0 mg/mL),
bacitracin (50 µg/mL), bestatin (30 µg/mL), captopril (10 µM),
and phenylmethanesulfonyl fluoride (0.1 mM) was used.
Assays were done in duplicate with 10 µM naltrexone hydro-
chloride to define nonspecific tissue binding. The binding
reaction was terminated by rapid filtration through presoaked
(0.5% poly(ethylenimine) solution) GF/B Whatman glass fiber
strips with a Brandel cell harvester followed immediately by
three rapid washes with 4-mL aliquots of ice-cold saline
solution. The filters were removed and soaked in 10 mL of
scintillation fluid at 4 °C for at least 6 h before bound
radioactivity was measured. The data were analyzed by
nonlinear least-squares regression analysis.
(2) (a) Hruby, V. J . Conformational and Topographical Consider-
ations in the Design of Biologically Active Peptides. Biopolymers
1993, 33, 1073-1082. (b) Hruby, V. J .; Al-Obeidi, F.; Kazmierski,
W. M. Emerging Approaches in the Molecular Design of Receptor
Selective Peptide Ligands: Conformational, Topographical and
Dynamic Considerations. Biochem. J . 1990, 268, 249-262.
(3) Porreca, F.; Bilsky, E. J .; Lai, J . Pharmacological Characteriza-
tion of Opioid δ- and κ-Receptors. In The Pharmacology of Opioid
Peptides; Tseng, T. F., Ed.; Harwood Academic Publishers: New
York, 1995; pp 219-238.
(4) (a) Hruby, V. J .; Gehrig, C. A. Recent Developments in Design
of Receptor Specific Opioid Peptides. Med. Res. Rev. 1989, 9,
343-401 (b) Hruby, V. J .; Kao, L. F.; Pettitt, B. M.; Karplus, M.
The Conformational Properties of the Delta Opioid Peptide
[D-Pen2,D-Pen5]enkephalin in Aqueous Solution Determined by
NMR and Energy Minimization Calculation. J . Am. Chem. Soc.
1988, 110, 3351-3359. (c) Nikiforovich, G. V.; Hruby, V. J .;
Prakash, O.; Gehrig, C. A. Topographical Requirements for
δ-Selective Opioid Peptides. Biopolymers 1991, 31, 941-955. (d)
Zimmerman, D. M.; Leander, J . D. Selective Opioid Receptor
Agonists and Antagonists: Research Tools and Potential Thera-
peutic Agents. J . Med. Chem. 1990, 33, 895-902. (e) Mosberg,
H. I.; Omnas, J . R.; Lomize, A.; Heyl, D. L.; Norde, I.; Mousigian,
C.; Davis, P.; Porreca, F. Development of a Model for the δ-opioid
Receptor Pharmacophore. 2. Conformationally Restricted Phe3
Replacements in the Cyclic δ-Receptor Selective Tetrapeptide
Tyr-c[D-Cys-Phe-D-Pen]-OH (J OM-13). J . Med. Chem. 1994, 37,
4384-4391.
In Vitr o Bioa ssa ys. Electrically induced smooth muscle
contractions from mouse vas deferens (MVD) and guinea pig
ileum (GPI) longitudinal muscle-myenteric plexus were used
for bioassays.29 Tissues came from male ICR mice weighing
25-30 g and from male Hartley guinea pigs weighing 150-
400 g. The tissues were tied to gold chains with suture silk,
suspended in 20-mL baths containing 37 °C oxygenated (95%
O2, 5% CO2) Krebs bicarbonate solution (magnesium-free for
the MVD), and allowed to equilibrate for 15 min. The tissues
were then stretched to optimal length previously determined
to be 1 g tension (0.5 g for MVD) and allowed to equilibrate
for 15 min. The tissues were stimulated transmurally between
platinum plate electrodes at 0.1 Hz for 0.4-ms pulses (2.0-ms
pulses for MVD) and supramaximal voltage. Drugs were
added to the baths in 20-60-µL volumes. The agonists
remained in tissue baths for 3 min and were removed by
rinsing several times with fresh Krebs solution. Tissues were
given 8 min to reequilibrate and regain predrug contraction
height. Antagonists were added to the bath 2 min prior to
the addition of the agonists. Percent inhibition was calculated
by dividing height for 1 min preceding the addition of the
agonist by the contraction height 3 min after exposure to the
agonist. The IC50 values represent the mean of not less than
four tissues. Relative potency estimates were determined by
fitting the mean data to the Hill equation by using a comput-
erized nonlinear least-squares method. In some cases, the
weak µ agonist action of these analogues did not permit
completion of full dose-response curves in the GPI.
(5) Knapp, R. J .; Santoro, G.; De Leon, I. A.; Lee, K. B.; Edsall, S.
A.; Waite, S.; Malatynska, E.; Varga, E.; Calderon, S. N.; Rice,
K. N.; Rothman, R. B.; Porreca, F.; Roeske, W. R.; Yamamura,
H. Y. Structure-Activity Relatioships for SNC80 and Related
Compounds at Cloned Human Delta and Mu Opioid Receptors.
J . Pharmacol. Exp. Ther. 1996, 277, 1284-1291.
(6) (a) Knapp, R. J .; Landsman, R.; Waite, S.; Malatynska, E.; Varga,
E.; Haq, W.; Hruby, V. J .; Roeske, W. R.; Nagase, H.; Yamamura,
H. Y. Properties of TAN-67, a Nonpeptidic δ-Opioid Receptor
Agonist, at Cloned Human δ- and µ-Opioid Receptors. Eur. J .
Pharmacol. 1995, 291, 129-134 (b) Portoghese, P. S.; Moe, S.
T.; Takemori, A. E. A Selective δ1 Opioid Receptor Agonist
Derived from Oxymorphone. Evidence for Separate Recognition
Sites for δ1 Opioid Receptor Agonists and Antagonists. J . Med.
Chem. 1993, 36, 2572-2574.
(7) Suzuki, T.; Tsuji, M.; Mori, T.; Misawa, M.; Endoh, T.; Nagase,
H. Effects of A Highly Selective Nonpeptide δ Opioid Receptor
Agonist, TAN-67, on Morphine-Induced Antinociception in Mice.
Life Sci. 1995, 57, 155-168.
(8) Li, X.; Knapp, R. J .; Stropova, D.; Varga, E.; Wang, Y.; Malatyn-
ska, E.; Calderon, S. N.; Rice, K.; Rothman, R.; Porreca, F.;
Hruby, V. J .; Roeske, W. R.; Yamamura, H. I. Trp284 of the
Human δ Opioid Receptor is Required for SNC121 Binding But
Not Binding of pCl-DPDPE, Deltorphin II or Naltrindole.
Analgesia 1995, 1, 539-542.
(9) Qian, X.; Liao, S.; Stropova, D.; Yamamura, H. I.; Hruby, V. J .
Novel Scaffolds for Non-Peptide Mimetics of δ Opioid Receptor
Agonists Based on Peptide Leads. Regul. Pept. 1996, 65, 79-
82.
(10) Qian, X.; Shenderovich, M. D.; Ko¨ve´r, K. E.; Davis, P.; Horva´th,
R.; Zalewska, T.; Yamamura, H. I.; Porreca, F.; Hruby, V. J .
Probing the Stereochemical Requirements for Receptor Recogni-
tion of δ Opioid Agonists Through Topographic Modifications
in Position 1. J . Am. Chem. Soc. 1996, 118, 7280-7290.
(11) Shenderovich, M. D.; Liao, S.; Qian, X.; Hruby, V. J . Three-
Dimensional Models of δ Opioid Pharmacophore: Molecular
Modeling and QSAR Study of Peptide and Non-Peptide Ligands.
In Peptides: Chemistry, Structure and Biology, Proceedings of
the 15th American Peptide Symposium; Tam, J . P., Kaumaya,
P. T. P., Eds.; ESCOM Kluwer Acad. Publ.: Dordrecht, The
Netherlands, 1998, in press.
(12) Mosberg, H. I.; Hurst, R.; Hruby, V. J .; Gee, K.; Yamamura, H.
I.; Galligan, J . J .; Burks, T. F. Bis-Penicillamine Enkephalins
Possess Highly Improved Specificity Toward Delta Opioid
Receptors. Proc. Natl. Acad. Sci. U.S.A. 1983, 80, 5871-5874.
(13) (a) Weber, S. J .; Greene, D. L.; Hruby, V. J .; Yamamura, H. I.;
Porreca, F.; Davis, T. P. Whole Body and Brain Distribution of
[3H]Cyclic[D-Pen2,D-Pen5]Enkephalin After Intraperitoneal, In-
travenous, Oral and Subcutaneous Administration. J . Pharma-
col. Exp. Ther. 1992, 263, 1308-1316. (b) Williams, S. A.;
Abbruscato, J . J .; Hruby, V. J .; Davis, T. P. Passage of a δ-Opioid
Receptor Selective Enkephalin, [D-Penicillamin2,5]Enkephalin,
Across the Blood-Brain and the Blood-Cerebrospinal Fluid
Barriers. J . Neurochem. 1996, 66, 1289-1299.
Ack n ow led gm en t. The authors wish to acknowl-
edge the financial support of grants from NIDA
(U.S.P.H.S. DA08657 and DA06284), the Dean’s Fel-
lowship from the Graduate College of The University
of Arizona for S.L., and CONACyT-Mexico for J .A.L. The
results are those of the authors and do not necessarily
represent the official views of the U.S. Public Health
Service.
Refer en ces
(1) (a) Hirschmann, R. F. Medicinal Chemistry in the Golden Age
of Biology. Lessons from Stereoid and Peptide Research. Angew.
Chem., Int. Ed. Engl. 1991, 30, 1278-1301. (b) Moore, G. J .
Design Peptide Mimetics. Trends Pharmacol. Sci. 1994, 15, 124-
129. (c) Giannis, A.; Kolter, T. Peptidomimetics for Receptor
Ligands Discovery, Development and Medical Perspectives.
Angew. Chem., Int. Ed. Engl. 1993, 32, 1244-1267. (d) Sawyer,
T. K. Peptidomimetic Design and Chemical Approaches to

4776 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 24
Liao et al.
(14) Nikiforovich, G. V.; Prakash, O.; Gehrig, C. A.; Hruby, V. J .
Solution Conformations of the Peptide Backbone for DPDPE and
Its â-Me-Phe4-Substituted Analogues. Int. J . Pept. Protein Res.
1993, 41, 347-361.
(15) Flippen-Anderson, J . L.; Hruby, V. J .; Collins, N.; George, C.;
Cudney, B. X-ray Structure of [D-Pen2,D-Pen5]Enkephalin, A
Highly Potent, Delta Opioid Receptor Selective Compound:
Comparisons with Proposed Solution Conformations. J . Am.
Chem. Soc. 1994, 116, 7523-7531.
(16) Hruby, V. J .; To¨th, G.; Gehrig, C. A.; Kao, L.-F.; Knapp, R.; Lui,
G. K.; Yamamura, H. I.; Kramer, T. H.; Davis, P.; Burks, T. F.
Topographically Designed Analogues of [D-Pen2,D-Pen5]-
enkephalin. J . Med. Chem. 1991, 34, 1823-1830.
Human δ-Opioid Receptor Determines the Selectivity of δ-Opioid
Agonist. Mol. Pharmacol. 1996, 50, 1619-1624.
(23) Quock, R. M.; Hosohata, Y.; Knapp, R. J .; Burkey, T. H.;
Hosohata, K.; Zhang, X.; Rice, K. C.; Nagase, H.; Hruby, V. J .;
Porreca, F.; Roseke, W. R.; Yamamura, H. I. Relative Efficacies
of δ-Opioid Receptor Agonists at the Cloned Human δ-Opioid
Receptor. Eur. J . Pharmacol. 1997, 326, 101-104.
(24) Gisin, B. F. The Preparation of Merrifield-Resins Through Total
Esterification With Cesium Salts. Helv. Chim. Acta 1973, 56,
1476-1482.
(25) Misicka, A.; Hruby, V. J . Optimization of Disulfide Bond
Formation. Polish J . Chem. 1994, 68, 893-899.
(17) Mosberg, H. I.; Hurst, R.; Hruby, V. J .; Gee, K.; Akiyama, K.;
Yamamura, H. I.; Galligan, J . J .; Burks, T. F. Cyclic Penicil-
lamine Containing Enkephalin Analogues Display Profound
Delta Receptor Selectivities. Life Sci. 1984, 33, 447-450.
(18) Nikiforovich, G. V. Towards Nonpeptide Agonists: Design of
“True” Peptidomimetics. Lett. Pept. Sci. 1995, 2, 172-176.
(19) Farmer, P. S. Binding the Gap Between Bioactives Peptides and
Non-Peptides: Some Perspectives in Design. In Drug Design;
Ariens, E. J ., Ed.; Academic Press: New York, 1980; Vol. X, p
127.
(20) Corey, E. J .; Gras J .-L.; Ulrich, P. A New General Method for
Protection of the Hydroxyl Functions. Tetrahedron Lett. 1976,
809-812.
(21) Lee, J . B.; Nolan, T. J . Sugar Esters III. New Reagent for
Deoxyhalo Sugar Derivatives. Can. J . Chem. 1966, 44, 1331-
1334.
(26) Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. J . Color
Test for Detection of Free Terminal Amino Groups in the Solid-
Phase Synthesis of Peptides. Anal. Biochem. 1970, 34, 595-598.
(27) Vaughn, L. K.; Knapp, R. J .; Toth, G.; Wan, Y.-P.; Hruby, V. J .;
Yamamura, H. I. A Highly Affinity, Highly Selective Ligand for
the Delta Opioid Receptor: [3H]-[D-Pen2,pCl-Phe4,D-Pen5]-
Enkephalin. Life Sci. 1989, 45, 1001-1008.
(28) Rothman, R. B.; Xu, H.; Seggel, M.; J acobson, A. E.; Rice, K. C.;
Brine, G. A.; Carroll, F. I. RTI-4614-4: An Analogue of (+)-Cis-
3-methylfentanyl with a 27,000-Fold Binding Selectivity for Mu
Versus Delta Opioid Binding Sites. Life Sci. 1991, 48, PL111-
116.
(29) Kramer, T. H.; Davis, P.; Hruby, V. J .; Burks, T. F.; Porreca, F.
In Vitro Potency, Affinity and Agonist Efficacy of Highly
Selective Delta Opioid Receptor Ligands. J . Pharmacol. Exp.
Ther. 1993, 266, 577-584.
(22) Varga, A. E.; Li, X.; Stropova, D.; Zalewska, T.; Landsman, R.
S.; Knapp, R. J .; Malatynska, E.; Kawai, K.; Mizusura, A.;
Nagase, H.; Calderon, S. N.; Rice, K. C.; Hruby, V. J .; Roeske,
W. R.; Yamamura, H. I. The Third Extracellular Loop of the
J M980374R