Synthesis, biological evaluation, and automated docking of constrained analogues of the opioid peptide H-Dmt-d-Ala-Phe-Gly-NH2 using the 4- or 5-methyl substituted 4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-one scaffold

Article abstract of DOI:10.1021/jm2003574

The Phe³ residue of the N-terminal tetrapeptide of dermorphin (H-Dmt-d-Ala-Phe-Gly-NH₂) was conformationally constrained using 4- or 5-methyl-substituted 4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-one (Aba) stereoisomeric scaffolds. Several of the synthesized peptides were determined to be high affinity agonists for the μ opioid receptor (OPRM) with selectivity over the δ opioid receptor (OPRD). Interesting effects of the Aba configuration on ligand binding affinity were observed. H-Dmt-d-Ala-erythro-(4S, 5S)-5-Me-Aba-Gly-NH₂9 and H-Dmt-threo-(4R,5S)-5-Me-Aba-Gly-NH ₂12 exhibited subnanomolar affinity for OPRM, while they possess an opposite absolute configuration at position 4 of the Aba ring. However, in the 4-methyl substituted analogues, H-Dmt-d-Ala-(4R)-Me-Aba-Gly-NH₂14 was significantly more potent than the (4S)-derivative 13. These unexpected results were rationalized using the binding poses predicted by molecular docking simulations. Interestingly, H-Dmt-d-Ala-(4R)-Me-Aba-Gly-NH₂14 is proposed to bind in a different mode compared with the other analogues. Moreover, in contrast to Ac-4-Me-Aba-NH-Me, which adopts a β-turn in solution and in the crystal structure, the binding mode of this analogue suggests an alternative receptor-bound conformation.

Full text of DOI:10.1021/jm2003574

ARTICLE
pubs.acs.org/jmc
Synthesis, Biological Evaluation, and Automated Docking of
Constrained Analogues of the Opioid Peptide H-Dmt-D-Ala-Phe-Gly-
NH2 Using the 4- or 5-Methyl Substituted 4-Amino-1,2,4,
5-tetrahydro-2-benzazepin-3-one Scaffold
Rien De Wachter,† Chris de Graaf,*,‡,§ Atilla Keresztes,|| Bart Vandormael,† Steven Ballet,† Gꢀeza Tꢀoth,||
Didier Rognan,‡ and Dirk Tourwꢀe*,†
†Department of Organic Chemistry, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium
‡Structural Chemogenomics, UMR 7200 CNRS-UdS, Universitꢀe de Strasbourg, Illkirch F-67401, France
§Leiden/Amsterdam Center for Drug Research, Division of Medicinal Chemistry, Faculty of Science, VU University Amsterdam,
Amsterdam, The Netherlands
Institute of Biochemistry, Biological Research Centre, Hungarian Academy of Sciences, Szeged, Hungary
S Supporting Information
b
ABSTRACT: The Phe3 residue of the N-terminal tetrapeptide of dermorphin (H-Dmt-D-Ala-Phe-Gly-
NH2) was conformationally constrained using 4- or 5-methyl-substituted 4-amino-1,2,4,5-tetrahydro-2-
benzazepin-3-one (Aba) stereoisomeric scaffolds. Several of the synthesized peptides were determined to
be high affinity agonists for the μ opioid receptor (OPRM) with selectivity over the δ opioid receptor
(OPRD). Interesting effects of the Aba configuration on ligand binding affinity were observed. H-Dmt-D-
Ala-erythro-(4S,5S)-5-Me-Aba-Gly-NH2 9 and H-Dmt-threo-(4R,5S)-5-Me-Aba-Gly-NH2 12 exhibited
subnanomolar affinity for OPRM, while they possess an opposite absolute configuration at position 4 of
the Aba ring. However, in the 4-methyl substituted analogues, H-Dmt-D-Ala-(4R)-Me-Aba-Gly-NH2 14
was significantly more potent than the (4S)-derivative 13. These unexpected results were rationalized
using the binding poses predicted by molecular docking simulations. Interestingly, H-Dmt-D-Ala-(4R)-
Me-Aba-Gly-NH2 14 is proposed to bind in a different mode compared with the other analogues.
Moreover, in contrast to Ac-4-Me-Aba-NH-Me, which adopts a β-turn in solution and in the crystal
structure, the binding mode of this analogue suggests an alternative receptor-bound conformation.
’ INTRODUCTION
high sequence homology in their TM (73ꢀ76%) and intracellular
domains (63ꢀ66%). A large divergence exists in the N- and C-terminal
domains and in the extracellular loops (34ꢀ40% identity).19,20
The use of X-ray diffraction analysis or NMR spectroscopy to
directly evaluate receptorꢀligand interactions is complicated by
the fact that GPCRs are embedded in the cell membrane.21 So
far, crystal structures of rhodopsin (Rho)4 and more recently
opsin (Ops),11 β-1 (ADRB1) and β-2 (ADRB2) adrenergic,7ꢀ10
dopamine D3 (DRD3),22 A2A adenosine,12 and CXCR4
chemokine23 receptors have been reported. For all other GPCRs,
however, such crystallographic data are still absent and indirect
experimental methods have to be applied in combination with in
silico modeling techniques to locate key receptorꢀligand interactions
and construct three-dimensional receptorꢀligand models.24,25
In this way, site-directed mutagenesis studies, chimeric receptor
studies, incorporation of metal binding sites, and analysis of
ligand structureꢀaffinity relationships have previously been used
to investigate opioid receptorꢀligand interactions.16,26ꢀ33
Opioid receptors are involved in pain control, regulation of
mood, reward, motivation, and response to stress. Four different
types of opioid receptors have been identified: μ (OPRM), δ
(OPRD), k (OPRK), and the nociceptin (OPRX) receptor.1
These receptors can be activated by endogeneous opioid pep-
tides or exogenously administered opiates, which can act as
analgesic drugs for severe pain. On the other hand, the binding of
these ligands to the opioid receptors can result in other effects
such as respiratory depression, euphoria, effects on mood, and
inhibition of the gastrointestinal transit causing severe constipa-
tion and addiction. As a consequence, a straightforward goal of
opiate research is the development of a drug lacking these
undesirable side effects.2
Opioid receptors belong to the rhodopsin subclass within the
superfamily of G-protein-coupled receptors (GPCRs).3 These
receptors are composed of a core domain of seven transmem-
brane α-helices (TM), connected by three intra- (IL) and three
extracellular (EL) loops, and contain glycosylated N-terminal
and palmitoylated C-terminal domains of variable lengths.3ꢀ12
In the early 1990s, the human OPRM,13 OPRD,14,15 and
OPRK16ꢀ18 were cloned. Subtype protein sequences demonstrate
Received: March 28, 2011
Published: August 29, 2011
r
2011 American Chemical Society
6538
dx.doi.org/10.1021/jm2003574 J. Med. Chem. 2011, 54, 6538–6547
|

Journal of Medicinal Chemistry
ARTICLE
Such in silico studies can provide valuable insight into the binding
mode of opioid ligands, in particular of flexible peptide ligands.26 The
use of conformational constraints to limit peptide flexibility has been
proposed as a method to obtain information about the receptor-
bound conformation from the preferred conformation,34ꢀ37 based
on the assumption that a change from the preferred conformation
upon receptor binding has become energetically unfavorable. How-
ever, energetically unfavorable conformational rearrangements can
be tolerated in some cases without penalizing the tightness of
proteinꢀligand binding.38 In the current study we have introduced
conformationally constrained Phe analogues into opioid peptides
and studied the binding mode of these analogues to the OPRM by
molecular docking simulations. We show that opioid tetrapeptides
with different (locally constrained) geometries can have a high
affinity for OPRM by adopting different bound conformations in
the receptor binding pocket. Interestingly, our studies suggest that
the docked receptor bound ligand conformation can deviate from the
low energy conformation of the unbound ligand as determined by
solution NMR.37
Among the naturally occurring opioids, the heptapeptide
dermorphin (H-Tyr-D-Ala-Phe-Gly-Tyr-Pro-Ser-NH2), which
was isolated from the skin of the South American frog Phyllome-
dusa sauvagei,39 is one of the most potent and selective μ-opioid
receptor agonists.40 The minimal sequence required for opiate-
like activity in vivo was shown to be the N-terminal tetra-
peptide.41,42 The replacement of Tyr1 in opioid peptides with
the unnatural amino acid 20,60-dimethyltyrosine (Dmt) increased
their μ- or δ-opioid potency by 1 or 2 orders of magnitude.43
H-Dmt-D-Ala-NH-(CH2)3-Ph was the first reported analogue
of this type and was proven to be a potent but only moder-
ately selective μ-agonist.44 Subsequently, Dmt substitution
yielded extraordinary potent and selective opioids such as
[Dmt1]DALDA (H-Dmt-D-Arg-Phe-Lys-NH2),45,46 [Dmt1]TAPP
(H-Dmt-D-Ala-Phe-Phe-NH2),47 and [Dmt1]TIPP-NH2 (H-Dmt-
Tic-Phe-Phe-NH2).48 The opioid ligand DIPP-NH2[Ψ] (H-
Dmt-Ticψ(CH2NH)Phe-Phe-NH2) is an opioid μ agonist/δ
antagonist that produces a potent analgesic effect, no physical
dependence, and less tolerance than morphine in rats.48
Figure 1. 4-Amino-1,2,4,5-tetrahydro-2-benzazepin-3-one 1 (Aba) and
8-hydroxy-4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-one
2 (Hba)
scaffolds.
20.8 nM) is the most potent agonist in the GPI (guinea pig ileum,
IC50 = 3.64 nM) assay which contains mainly μ-opioid receptors.
This peptide also shows potent in vivo antinociception after i.t. and
iv administration and in addition possesses a prolonged duration of
action.52,53
We have recently reported that β-methylation of the Aba
scaffold resulted in an unchanged ring and backbone conforma-
tion for both stereoisomers of (5S)- or (5R)-Me-Aba.37 In
contrast, α-methylation resulted in 4-Me-Aba which adopted a
different ring conformation and induced a β-turn conformation
in tetrapeptide models.37 We have now investigated the effects of
a combined introduction of the conformationally constrained
Aba residue and of a methyl substituent in its 4- or 5-position in
the opioid peptide H-Dmt-D-Ala-Phe-Gly-NH2. A rationalization
of the biological data by using the binding poses proposed by
molecular docking simulations in OPRM will be presented.
’ RESULTS AND DISCUSSION
Synthesis. The synthesis of the Boc-protected erythro-(4S,5S)
3, erythro-(4R,5R) 4, threo-(4S,5R) 5, and threo-(4R,5S)-Me-
Aba-Gly-OH 6 building blocks (Figure 2) was performed as
described for the racemates37 using the corresponding enantio-
pure β-methylphenylalanine starting materials. Racemic erythro-
β-MePhe and racemic threo-β-MePhe were N-Cbz protected and
resolved by crystallization using quinine or quinidine according
to the procedure of Kataoka.54 Optical purity was determined after
N-deprotection by chiral derivatization with Marfey’s reagent
Nα-(2,4-dinitro-5-fluorophenyl)-L-alaninamide (FDAA),55 fol-
lowed by HPLC analysis, which also allowed the confirmation
of the absolute configuration of the samples.56 After the forma-
tion of the benzazepinone ring,37 the 5-methyl-Aba-Gly stereo-
isomers were N-Boc protected using Boc2O and the resulting
building blocks 3ꢀ6 were purified by crystallization as their
dicyclohexylamine salts. In contrast to the erythro isomers 3 and
4, the threo isomers 5 and 6 substantially epimerized during this
procedure, which led to a contamination of 5 with 4 and of 6 with
3. Fortunately, after peptide synthesis using these mixtures, the
resulting diastereomeric peptides could be separated by reversed-
phase preparative HPLC, and their configuration could be
assigned by comparison with the peptides containing the en-
antiopure erythro isomers 3 and 4.
Conformational restrictions in the flexibility of the peptide
side chains have proven to cause shifts in selectivity and affinity.
The use of β-methylated amino acids to bias the population of
the χ1 torsion angle has been pioneered by Hruby. Important
effects on the receptor affinity, on selectivity, on the agonist or
antagonist character, and on the duration of action have been
observed in a variety of peptides,34,35 including opioids.49,50
Cyclization provides an alternative way to limit the flexibility of
the side chain. The 4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-
one scaffold 1 (Aba, Figure 1) allows the side chain orientation of
the Phe3 residue to be fixed into the trans (χ1 = 180°) or the
gauche(+) (χ1 = +60°) staggered conformations.36,51
δ
Dermorphin, a μ-selective peptide (IC50μ = 2.4 nM, IC50
=
295 nM),51 loses its selectivity on incorporation of the confor-
mationally constrained dipeptide [Aba-Gly] in positions 3 and 4,
μ
δ
mainly by an increase in δ-affinity (IC50 = 11.0 nM, IC50
=
In contrast to the 5-Me-substituted Aba-Gly stereoisomers,
racemic H-(4R,S)-Me-Aba-Gly-OH was prepared using an ear-
lier published strategy.37 Briefly, racemic 2-amino-3-(2-cyano-
phenyl)-2-methylpropanoic acid was methyloxycarbonyl (Moc)
protected, and the nitrile was reduced with H2/10% Pd on C to
provided the o-aminomethyl compound. Intramolecular cycliza-
tion using 1-ethyl-3-(30-dimethylaminopropyl)carbodiimide (EDC)
provided (R,S)-4-methyloxycarbonylamino-4-methyl-1,2,4,5- tetra-
hydro-2-benzazepin-3-one which was N-alkylated with tert-butyl
17.4 nM).51 Introduction of the Hba 2 and Aba 1 scaffolds in
positions 1 and 3, respectively, of the N-terminal tetrapeptides
(H-Tyr-D-Ala-Phe-Gly-NH2) of dermorphin does not dramati-
cally change the μ affinity.52 The δ affinity on the other hand is
increased approximately by a factor of 100. Hence, the confor-
mationally restricted tetrapeptides have a selectivity shifted
toward OPRD. In a functional assay, however, the double constrained
μ
carboxamide tetrapeptide H-[Hba-D-Ala]-[Aba-Gly]-NH2 (IC50
=
6539
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547

Journal of Medicinal Chemistry
ARTICLE
Figure 2. Boc-protected erythro-3 (4S,5S) and 4 (4R,5R) and threo-5-Me-Aba-Gly-OH 5 (4S,5R) and 6 (4R,5S) and 4-Me-Aba-Gly-OH 7 (4S) and 8
(4R) building blocks.
consistent with the role of Dmt in opioid peptide analogues.
Analogue 9 shows a very high μ-affinity and also a good δ-affinity,
both of which are comparable to that of the unsubstituted
reference compound 15. Interestingly, analogue 12, which has
the threo-(4R,5S) configuration, exhibits a similar high affinity for
OPRM, although it possesses an opposite absolute configuration
at position 4 of the Aba ring. This is unusual, since in opioid
Table 1. Constrained Tetrapeptide Analogues 9ꢀ16
H-Dmt-D-Ala-erythro-(4S,5S)-5-Me-Aba-Gly-NH2 9
H-Dmt-D-Ala-erythro-(4R,5R)-5-Me-Aba-Gly-NH2 10
H-Dmt-D-Ala-threo-(4S,5R)-5-Me-Aba-Gly-NH2 11
H-Dmt-D-Ala-threo-(4R,5S)-5-Me-Aba-Gly-NH2 12
H-Dmt-D-Ala-(4S)-Me-Aba-Gly-NH2 13
H-Dmt-D-Ala-(4R)-Me-Aba-Gly-NH2 14
H-Dmt-D-Ala-(4S)-Aba-Gly-NH2 15
H-Dmt-D-Ala-(4R)-Aba-Gly-NH2 16
peptides a D-configuration of residue 3 results in a decreased
potency,49,50,60,61 as is also the case for the (4R)-Aba-containing
analogue 16, compared to its (4S)-epimer 15. A similar unchar-
acteristic observation was made for the α-methyl substituted
analogues, for which compound 14, having a (4R)-configuration
at the α-carbon, is more potent than its (4S)-derivative 13. The
former of these has a similar affinity for the OPRM (Kiμ = 0.3 nM)
to the erythro-(4R,5R)-5-methyl isomer 10 (Kiμ = 0.4 nM).
Compound 10 is, however, more selective. All compounds show
agonist properties in the [35S]GTPγS-binding experiments with
good potencies except for compounds 11 and 13, which can be
considered as low potency agonists, in agreement with their
lower binding affinities.
Rationalization of the Biological Data by Molecular Dock-
ing Simulations to a OPRM Receptor Model. An attempt to
explain the unexpected effects of the configuration at C-4 of the
Aba residue in the tetrapeptide analogues was made by con-
structing a model for the OPRM and observing the binding poses
of the ligands after automated docking. The recently solved
ADRB2 crystal structure (PDB code 2RH1.pdb)8 was used as a
template to model the TM helices of the preliminary OPRM
model, with the exception of TM2, which was derived from a
previously reported CCR5 model62 (the OPRM model was
constructed before the release of the recently solved CXCR4
chemokine).23 Arguments for this modeling template selection
procedure are (i) the larger distance between TM3 and TM6 at
the bottom of the ligand binding pocket in ADRB2 compared to
the bovine rhodopsin crystal structure,4,8 offering more space for
docking the Dmt scaffold of the peptide ligands and (ii) the
T2.56XP2.58 motif present in OPRM. This motif has been
suggested to induce an alternative kink in TM263,64 compared
to the rhodopsin and β adrenergic receptor crystal structures4,8,10
and has been carefully modeled in the CCR5 template.62 An extra
turn of TM6 (6.56ꢀ6.58), derived from the ADRB2 crystal
structure, was added to the preliminary model, locating lysine at
position 6.58, which plays an important role in determining
opioid receptor specificity,65 into the binding pocket. The
remainder of the procedure is described in the Experimental
Section.
bromoacetate and NaH as a base to Moc-(4R,S)-Me-Aba-Gly-
OtBu. After Moc deprotection and ester hydrolysis in a 33%
solution of HBr in AcOH, the compound was Boc-protected to
Boc-(4R,S)-Me-Aba-Gly-OH. After incorporation into the tetra-
peptide sequence the two epimeric peptides were separated by
HPLC. To discriminate the (4S)- from the (4R)-isomer, an
asymmetric synthesis of Boc-(4S)-Me-Aba-Gly-OH 7 (Figure 2)
was required, which, after incorporation into the tetrapeptide,
allowed the assignment of the (4S)-epimer by comparison of the
HPLC retention times. Therefore, a small scale synthesis of (S)-
α-methyl-o-cyanophenylalanine was performed by alkylation of
the Sch€ollkopf chiral bislactim ether (2R,5SR)-2-isopropyl-
3,6-dimethoxy-5-methyl-2,5-dihydropyrazine with o-cyanoben-
zyl bromide57,58 and further transformation as described above
for the racemate.37
Peptide Synthesis. Analogues 9ꢀ16 of the tetrapeptide
H-Dmt-D-Ala-Phe-Gly-NH2, containing the constrained build-
ing blocks 3ꢀ8 (Table 1), were prepared by solid-phase synth-
esis using Boc-chemistry and the 4-methylbenzhydrylamine resin
as the solid support. Analogues 15 and 16 were prepared as reference
ligands using Boc-(4S)-Aba-Gly-OH and Boc-(4R)-Aba-Gly-
OH, respectively.34 The target peptides were cleaved from the
resin by HFliq and purified by preparative high-performance
liquid chromatography to a purity g95%.
Biological Evaluation: Receptor Binding Affinity and
Functional Bioactivity. Receptor binding affinities of com-
pounds 9ꢀ16 for OPRM and OPRD were determined by
displacement of [3H]DAMGO and [3H][Ile5,6]deltorphin-2
from rat brain membrane binding sites, respectively (Table 2).
The in vitro activity of the ligands was tested in [35S]GTPγS-
binding experiments in rat brain membranes.59
The data in Table 2 indicate that all compounds have a higher
affinity for OPRM than for OPRD. The affinities are much higher
than those of the tyrosine-containing analogues,52,53 which is
6540
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547

Journal of Medicinal Chemistry
ARTICLE
Table 2. Binding Affinities and Functional Activities of Tetrapeptides 9ꢀ16
[
35S]GTPγS
[3H]DAMGO,
Kiμ (nM)
[3H]Ile5,6
-deltorphin-2, Kiδ (nM)
selectivity
Kiδ/Kiμ (nM)
peptide
EC50(nM)
Emax (%)
H-Dmt-D-Ala-[erythro-(4S,5S)-5-Me-Aba-Gly]-NH2 9
H-Dmt-D-Ala-[erythro-(4R,5R)-5-Me-Aba-Gly]-NH2 10
H-Dmt-D-Ala-[threo-(4S,5R)-5-Me-Aba-Gly]-NH2 11
H-Dmt-D-Ala-[threo-(4R,5S)-5-Me-Aba-Gly]-NH2 12
H-Dmt-D-Ala-(4S)-4-Me-Aba-Gly-NH2 13
0.025 ( 0.012
0.4 ( 0.08
1.7 ( 0.9
490 ( 71
720 ( 109
24 ( 6.4
859 ( 107
63 ( 1.5
2.4 ( 0.6
337 ( 28
68
1225
51
36 ( 6
191 ( 11
167 ( 3
184 ( 46
685 ( 114
9.3 ( 1.3
941 ( 114
288 ( 44
18 ( 6
14 ( 1.5
130 ( 3
0.025 ( 0.002
4.6 ( 1.7
960
187
210
51
149 ( 3
159 ( 1.2
154 ( 10
161 ( 15
178.2 ( 3.5
H-Dmt-D-Ala-(4R)-4-Me-Aba-Gly-NH2 14
H-Dmt-D-Ala-(4S)-Aba-Gly-NH2 15
0.3 ( 0.08
0.047 ( 0.012
0.33 ( 0.06
H-Dmt-D-Ala-(4R)-Aba-Gly-NH2 16
1021
64.2 ( 2.1
12 and medium affinity compound 14 form the same H-bond
interactions with D3.32 and H6.52 as the N-terminal tyramine
group of 17 and bind in the same hydrophobic pocket between
Y3.33, F5.43, F5.47, and W6.48. The protonated nitrogen of
medium affinity compound 11 is just out of H-bond distance
(but within ionic interaction distance) to D3.32 but forms an
H-bond with H6.52 as well (Figure 4C,E). The Aba group of high
affinity compounds 9 and 12 stacks with W7.35 and H7.36,
residues that play an important role in OPRM-specific agonist
binding67 (Figure 4A,B,D), and occupies the same binding
pocket as the Phe3 ring of 17 between the top of TM7 and the
second (G45.46 and F45.54) and third (T6.70) extracellular
loops.26 The 5-methyl group of 9 and 12 binds close to I7.39, a
residue position critical for the selectivity of nociceptin receptor
(T7.39 in wild-type) against opioid ligands,70 and this interaction
might be responsible for the increased affinity compared to the
nonmethylated reference compound 15. The C-terminal amide
oxygen atom of these compounds forms an H-bond with the
OPRM-specific residue K6.58 (like 17). These data indicate that
despite a different configuration at C-4 of the Aba ring, analogues
9 and 12 can adopt very similar binding poses, explaining their
high affinity for the OPRM. For medium affinity compound 14 a
different binding mode than for compounds 9, 11, and 12 was
observed in which the Aba ring is stacking with the Dmt ring and
is occupying the same space as the S-CH2-CH2-S bridge of 17
(Figure 4D). Interestingly, compound 14 is assumed to adopt a
β-turn in solution by forming an intramolecular H-bond between
the C-terminal carboxamide oxygen and the Ala backbone amide
nitrogen atom.37 Although compound 14 can adopt a β-turn in
conformational searches (in line with previous ligand-based
modeling studies,37 results not shown), this β-turn conformation
cannot be accommodated in the OPRM receptor model. In the
proposed binding mode the C-terminal carboxamide group of 14
forms an intermolecular H-bond with K6.58 instead (Figure 4D).
The predicted extended conformations of compounds 9, 11, and
12 are consistent with the NMR data on the preferred conforma-
tion of 5-Me-Aba tetrapeptide models.37
Figure 3. Tyr-c(S-Et-S)[D-Cys-Phe-D-Pen]NH2 17 (JOM-6).24
The starting point for the docking studies was the cyclic
tetrapeptide 17 (JOM-6, Tyr-c(S-Et-S)[D-Cys-Phe-D-Pen]NH2,
Figure 3), which is a highly potent and selective agonist for
OPRM (Ki = 0.17 nM) and for which binding poses in an OPRM
model have been reported.26
The binding mode of the highest ranked Surflex, version
2.301,66 docking pose of 17 in the OPRM receptor model
(docking procedure described in the Experimental Section) is
line with experimental data (Figure 4A).26,27 The protonated
amine group of the ligand forms a complementary H-bond
interaction network with D3.3228,29 and Y7.43,30,31 while the
Tyr1 phenol ring binds in the hydrophobic pocket between
Y3.3330,32 and W6.4832 and forms an H-bond with H6.52.33 The
Phe3 group of 17 adopts a trans orientation26 and forms an
aromatic interaction with W7.35 in OPRM.67 The C-terminal
carboxamide group of 17 is in proximity of E5.35 in OPRM.19,26
Compounds 9ꢀ14 were docked into the refined OPRM
receptor model with Surflex, version 2.301.66 The best pose of
17 was used to generate a reference interaction fingerprint (IFP)
as previously described,68 and a Tanimoto coefficient (Tc-IFP)
measuring IFP similarity with the reference 17 pose in the
OPRM receptor model was used to filter and rank the docking
poses of compounds 9ꢀ14.
Docking poses of 9, 11, 12, and 14 in the OPRM structure (see
Figure 4AꢀC) involved in the same receptorꢀligand interac-
tions as 1726 (Figure 4A) were selected using a receptorꢀligand
interaction fingerprint (IFP) scoring method68 as described
earlier.69 For high affinity compounds 9 and 12 and medium
affinity compound 14 (but not for medium affinity compounds
10, 11, and 13), poses could be found showing an ionic
interaction with D3.3228,29 as well as an H-bond to H6.52.33
Figure 4 shows the top-ranked docking poses of 9 (Figure 4B),
11 (Figure 4C), 12 (Figure 4AꢀC), and 14 (Figure 4D) as well
as a part of their receptorꢀligand interaction fingerprint (IFP)
bit strings (Figure 4E), compared to the experimentally
supported19,26,27 binding mode and IFP of 17 (Figure 4A,E).
The N-terminal Dmt groups of high-affinity compounds 9 and
For medium/low affinity compounds 10 and 13 no docking poses
were generated in which the Dmt group forms an ionic interaction
with D3.3228,29 and an H-bond with H6.52.33 Comparison of the
binding mode of 12 (Figure 4) and the ligand-based rapid overlay of
chemical structures (ROCS)71 of 10 and 13 with the docking pose of
12 (Figure 5) suggests that the conformations of the erythro-
(4R,5R)-5-Me-Aba (10) (Figure 5A) and (4S)-Me-Aba (13)
(Figure 5B) methyl groups do not properly match the shape of
the OPRM specific binding pocket between W7.35 and H7.36.67
6541
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547

Journal of Medicinal Chemistry
ARTICLE
Figure 4. Docking poses selected by IFP of compounds 12 (AꢀD, green carbon atoms), 9 (B, yellow), 11 (C, orange), and 14 (D, brown) in the OPRM
receptor model compared to the reference binding mode of 17 (A, magenta carbon atoms). Parts of the backbone of transmembrane (TM) helices 6 and
7 are represented by light yellow ribbons (TM3 and second (EL2) and third (EL3) extracellular loops are not shown for clarity). Important binding
residues are depicted as ball-and-sticks with gray carbon atoms. Oxygen, nitrogen, sulfur, and hydrogen atoms are colored red, blue, orange, and cyan,
respectively. H-bonds described in the text are depicted by black dots. The IFP bit strings of compounds 12 (AꢀD), 9 (B), 11 (C), and 14 (D) are
compared to the reference IFP of 17 (A) in part E. For reasons of clarity, the bit strings of only eight residues (out of 26) are shown as an example.
receptor binding pocket by adopting different conformations and
support the idea that OPRM bound ligand conformations can
deviate from low energy unbound ligand conformations.
’ EXPERIMENTAL SECTION
General Methods. RP-HPLC was performed using a RP C-18
column (Supelco Discovery BIO Wide Pore C18, l = 25 cm, d = 0.45 cm,
PS = 5 μm) with a mobile phase consisting of water/acetonitrile
containing 0.1% TFA. Products were eluted using the following para-
meters: gradient, t = 0 min, 3% CH3CN, t = 20 min, 97% CH3CN; flow
rate, 1.0 mL minꢀ1; λ = 215 nm. Preparative HPLC was performed using
Figure 5. ROCS alignments of compound 10 (pink carbon atoms, A)
and 13 (magenta, B) to the docking pose of compound 12 (green). The
rendering and visual orientation are the same as in Figure 4. Misaligned
methyl groups are indicated by the corresponding compound number.
a RP C-18 column (Supelco Discovery BIO Wide Pore C18, l = 25 cm,
d = 2.12 cm, PS = 10 μm) with the above-mentioned gradient at a flow
rate of 20.0 mL minꢀ1. TLC analysis was performed on plastic sheets
precoated with silica gel 60F254 (Merck). Melting points were measured
with a B€uchi B 540 melting point apparatus, with a temperature
increment of 1 °C minꢀ1. 1H and 13C NMR spectra were recorded at
250.13 and 62.90 MHz, respectively, with a Bruker Avance DRX250
spectrometer, using TMS or the residual solvent signal as internal
reference. For some advanced NMR analyses samples were measured
with the Bruker AMX500 spectrometer, using pulse sequences of the
Bruker program library. Mass spectra were recorded on a VG Quattro II
spectrometer using electrospray ionization (positive ion mode). The
MBHA resin was purchased from Neosystem (Strasbourg, France). Boc-
Dmt-OH was obtained from RSP Amino Acids (Shirley, MA). 2-1H-
(Benzotriazol-1-yl)-1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU)
was purchased from Senn Chemicals (Dielsdorf, Switzerland).
The structureꢀactivity relationships and molecular docking
studies of our conformationally restricted tetrapeptides indicate
that in addition to recognition of an aromatic amine and aromatic
moiety between TM3 (D3.32, Y3.33), TM6 (W6.48, H6.52),
and TM7 (Y7.43), a tight steric fit of the ligand in the hydro-
phobic binding pocket between W7.35 and H7.36 (two residues
that play an important role in OPRM-specific agonist binding67)
is essential for high OPRM affinity. We therefore propose
targeting this binding pocket with apolar moieties as a general
design strategy for OPRM ligands.
Synthetic Procedures. For the synthesis of Boc-protected
erythro-(4S,5S)- (3), erythro-(4R,5R)- (4), threo-(4S,5R)- (5), and
threo-(4R,5S)-5-Me-Aba-Gly-OH (6), see Supporting Information.
Peptide Synthesis. The N-terminal tetrapeptides 9ꢀ16 were
prepared through solid-phase synthesis using Boc chemistry and the
MBHA resin (0.95 mmol/g) as solid support. A 3-fold excess of the
amino acids and coupling reagent (TBTU) was used together with a
9-fold excess of the base NMM and a 1:1 mixture of dry DMF/dry
CH2Cl2 was used as a solvent system. Final cleavage of the peptide from
the resin was done by treatment with HFliq for 1 h at 0 °C (2 mL of
anisole and 10 mL of HFliq were applied for 1 g of peptideꢀresin). These
crude mixtures were purified by preparative HPLC, and the pure
’ CONCLUSIONS
Several of the tetrapeptides synthesized in this study are high
affinity OPRM agonist with selectivity over the OPRD. Interest-
ing effects of the Aba configuration on ligand binding affinity
were rationalized by molecular docking simulations of the opioid
tetrapeptides in a three-dimensional OPRM receptor model.
Compounds 9 and 12 have subnanomolar affinity for OPRM,
although they possess an opposite absolute configuration at position
4 of Aba, while compound 14 was significantly more potent than
its (4S)-stereoisomer 13. Our docking studies predict that high
affinity OPRM ligands can form similar interactions within the
6542
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547

Journal of Medicinal Chemistry
ARTICLE
peptides were isolated as their TFA salt with purities of g95% as
determined by analytical HPLC and TLC using the conditions men-
tioned above (see General Methods).
were washed with 3 ꢁ 5 mL of ice-cold Tris-HCl buffer (pH 7.4), and
then the filter-bound radioactivity was measured in an Optiphase
Supermix scintillation cocktail (Perkin-Elmer) by immersing the filters
in the cocktail, using a TRI-CARB 2100TR liquid scintillation counter.
Each experiment was performed in duplicate and analyzed by the one/
two-site binding competition fitting option of the GraphPad Prism
software.
H-Dmt-D-Ala-[erythro-(4S,5S)-5-Me-Aba-Gly]-NH2 TFA 9.
Preparative HPLC yielded the desired compound (white powder,
3
31%). HPLC (standard gradient): tR = 10.8 min. TLC Rf(EBAW) =
0.69. HRMS (ESP+) MW C27H35N5O5 TFA = 623.6 g/mol. Found m/z
3
510.2701 [M + H]+, C27H36N5O5 (without TFA) requires 510.2711.
H-Dmt-D-Ala-[erythro-(4R,5R)-5-Me-Aba-Gly]-NH2 TFA 10.
Ligand-Stimulated [35S]GTPγS Functional Assay. Rat brain
membranes (15 μg of protein/tube) were incubated with 0.05 nM
[35S]GTPγS and 10ꢀ10ꢀ10ꢀ5 M opioids in the presence of 30 μM
GDP, 100 mM NaCl, 3 mM MgCl2, and 1 mM EGTA in 50 mM Tris-
HCl pH 7.4 buffer for 60 min at 30 °C. Basal binding was measured in
the absence of opioids and was corrected for the nonspecific binding to
yield the specific binding. Nonspecific binding was determined with
10 μM unlabeled GTPγS. The reaction was initiated by the addition of
the protein and terminated by the addition of 5 mL of ice-cold buffer
(50 mM Tris-HCl, pH 7.4) to the vials and filtering the samples through
a Whatman GF/B glass fiber filter with a Brandel cell harvester. Vials
were washed three times with 5 mL of ice-cold buffer and then directly
immersed in an Optiphase Supermix scintillation cocktail. The radio-
activity was measured using a TRI-CARB 2100TR liquid scintillation
counter. Ligand stimulations were expressed as percentage of the specific
[35S]GTPγS binding over the basal activity. Each measurement was
performed in triplicate and analyzed by the sigmoid doseꢀresponse
curve fitting option of the GraphPad Prism software.
3
Preparative HPLC yielded the desired compound (white powder, 71%).
HPLC (standard gradient): tR = 10.3 min. TLC Rf(EBAW): 0. 54. HRMS
(ESP+) MW C27H35N5O5 TFA = 623.6 g/mol. Found m/z 510.2711
3
[M + H]+, C27H36N5O5 (without TFA) requires 510.2711.
H-Dmt-D-Ala-[threo-(4S,5R)-5-Me-Aba-Gly]-NH2 TFA 11. Pre-
3
parative HPLC yielded the desired compound (white powder, 23% + 19%:
erythro-(4R,5R)-derivative 10). HPLC (standard gradient): tR = 10.4 min.
TLC Rf(EBAW) = 0.70. HRMS (ESP+) MW C27H35N5O5 TFA =
3
623.6 g/mol. Found m/z 510.2705 [M + H]+, C27H36N5O5 (without
TFA) requires 510.2711.
H-Dmt-D-Ala-[threo-(4R,5S)-5-Me-Aba-Gly]-NH2 TFA 12. Pre-
3
parative HPLC yielded the desired compound (white powder, 22% + 28%
erythro-(4S,5S)-derivative 9). HPLC (standard gradient): tR = 10.9 min.
TLC Rf(EBAW): 0. 69. HRMS (ESP+) MW C27H35N5O5 TFA =
3
623.6 g/mol. Found m/z 510.2711 [M + H]+, C27H36N5O5 (without
TFA) requires 510.2711.
H-Dmt-D-Ala-[(4S)-Me-Aba-Gly]-NH2 TFA 13. Preparative
HPLC yielded the desired compound (white powder, 25%). HPLC
(standard gradient): tR = 10.3 min. TLC Rf(EBAW) = 0.71. HRMS
Homology Modeling and Automated Docking. In order to
get agonist-bound models of the OPRM receptor, we mainly followed a
previously defined five-step protocol.69
3
The recently solved ADRB2 crystal structure7ꢀ9 (PDB code 2RH1.pdb)8
was used as template to model the TM helices of the preliminary OPRM
model, with the exception of TM2, which was derived from a previously
validated CCR5 model.62 The BallesterosꢀWeinstein residue numbering24
scheme was used throughout this work for GPCR TM helices, while a
recently proposed numbering69 scheme was used to number the
residues in the second extracellular loop (EL2). An extra turn of TM6
(6.56ꢀ6.58), derived from the ADRB2 crystal structure, was added to
the preliminary model because position 6.58 plays an important role in
determining opioid receptor specificity.65
(ESP+) MW C27H35N5O5 TFA = 623.6 g/mol. Found m/z 510.2695
3
[M + H]+, C27H36N5O5 (without TFA) requires 510.2711.
H-Dmt-D-Ala-[(4R)-Me-Aba-Gly]-NH2 TFA 14. Preparative
3
HPLC yielded the desired compound (white powder, 21%). HPLC
(standard gradient): tR = 11.0 min. TLC Rf(EBAW) = 0.71. HRMS (ESP
+) MW C27H35N5O5 TFA = 623.6 g/mol. Found m/z 510.2720 [M +
3
H]+, C27H36N5O5 (without TFA) requires 510.2711.
H-Dmt-D-Ala-(4S)-Aba-Gly-NH2 TFA 15. Preparative HPLC
3
yielded the desired compound (white powder, 38%). HPLC (standard
gradient): tR = 10.1 min. TLC Rf(EBAW) = 0.65. HRMS (ESP+) MW
C26H33N5O5 TFA = 609.3 g/mol. Found m/z 496.2545 [M + H]+,
This preliminary high-throughput receptor model was generated
using the GPCRgen program.74 The amino acid sequence alignments
used for constructing the receptor models are shown in Supporting
Information Figure 1.
3
C26H34N5O5 (without TFA) requires 496.2554.
H-Dmt-D-Ala-(4R)-Aba-Gly-NH2 TFA 16. Preparative HPLC
3
yielded the desired compound (white powder, 41%). HPLC (standard
gradient): tR = 9.7 min. TLC Rf(EBAW) = 0.56. HRMS (ESP+) MW
In the second step, the known cyclic peptide agonist 17 was docked
into this preliminary model using Surflex, version 2.301.66 The trans ring
conformation of 1719,26,75 was chosen as input for the docking simula-
tions. Experimentally driven H-bond constraints were used to prelimi-
narily guide the docking process in the OPRM receptor (1) between the
protonated amine of Tyr1 of 17 and both D3.3228,29 carboxylate oxygen
atoms, (2) between the hydroxyl group of Tyr1 of 17 and the Nε2
hydrogen atom of H6.52,33 (3) between the C-terminal carboxamide
nitrogen of 17 and the carboxylate group of E5.35.19,26
In the third step, the agonistꢀreceptor complex was minimized with
AMBER 876 using the AMBER 03 force field to relax the structure and
remove steric bumps. The minimizations were performed by 1000 steps
of steepest descent followed by conjugate gradient until the rms gradient
C26H33N5O5 TFA = 609.3 g/mol. Found m/z 496.2550 [M + H]+,
3
C26H34N5O5 (without TFA) requires 496.2554.
Radioligand Binding Assay. For competitive binding experi-
ments, rat brain membrane homogenates (Wistar, male, 250ꢀ300 g
body weight) were used, which were prepared as described by Boz€u
et al.72 The protein content of the homogenates was determined by the
method of Bradford73 using BSA as a standard. The protein concentra-
tion was varied between 0.2 and 0.4 mg/test tube. The following
conditions were set to assess inhibitory constants: [3H]DAMGO
(25 °C, 1 h, GF/C filter, glass tube), [3H]Ile5,6deltorphin-2 (35 °C,
45 min, GF/B filter, plastic tube, 0.25 mg of BSA/tube). The incubation
mixtures were made up to a final volume of 1 mL with 50 mM Tris-HCl
buffer (pH 7.4), and samples were incubated in a shaking water bath at
the appropriate temperature. Competition binding experiments were
performed by incubating the membranes with 0.5 nM [3H]DAMGO, 2 nM
[3H]Ile5,6deltorphin-2, and increasing concentrations (10ꢀ11ꢀ10ꢀ5 M)
of unlabeled dermorphin analogues. Nonspecific binding was deter-
mined with 10 μM naloxone and subtracted from the total binding to
yield the specific binding. Incubation was initiated by the addition of the
membrane suspension and stopped by rapid filtration over Whatman
GF/C or GF/B glass fiber filters, using a Brandel cell harvester. Filters
of the potential energy was lower than 0.05 kcal/(mol Å). A twin cutoff
3
(12.0 and 15.0 Å) was used to calculate nonbonded electrostatic
interactions, and the nonbonded pair list was updated every 25 steps.
In the fourth step, a five-residue window around C45.50 was added to
the TM model by threading this part of EL2 onto the bRho crystal
structure and changing the residues in the respective residues of the
OPRM receptor. The rest of EL2 was constructed using two subsequent
Modeller 9v177 runs with explicit disulfide bridge constraints (between
C3.25 in TM3 and C45.50 in EL2) and including the agonist binding
6543
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547

Journal of Medicinal Chemistry
ARTICLE
pose in the TM template as “block” residue. In the first run, the bovine
rhodopsin crystal structure (PDB code 1U19.pdb)6 was used to model
the part upstream of EL2. Out of the 15 generated models, the model
with the highest Modeller and DOPE scores and EL2 loop conforma-
tions properly accommodating the original agonist binding orientation
in the original TM model was selected as input for a second Modeller
run. In this second run, the EL2 segment downstream from the β4 sheet
was constructed. One out of 15 models was again selected based on the
criteria described before. For the modeling of the EL2 of OPRM,
additional upper-bound distance constraints, based on experimental
data,26 were defined between the OD1 atom of D45.49 and the ND1
atom of H7.35 (3.5 Å) and the Phe3 of 17 and the CA atom of G45.46
(5.0 Å).
ionic interaction with D3.32 and H-bond interaction with H6.52 are
considered) and rank the docking poses of compounds 9ꢀ14. For
compounds 10 and 13 no docking poses were generated in which the
Dmt group forms an ionic interaction with D3.3228,29 and an H-bond
with H6.52.33 These compounds were therefore aligned with the selected
docking pose of compound 12 using ROCS, version 2.3.1.71 The
conformer databases of compounds 10 and 13 were generated using
standard settings of OMEGA, version 2.2.1,79 and the best alignments of
10 and 13 with 12 were selected based on Comboscore (combination of
shape Tanimoto and the normalized color score in this optimized overlay).71
’ ASSOCIATED CONTENT
After optimization of the EL2 conformation, extracellular loops 1 and
3, intracellular loops 1, 2, and 3, and helix 8 were modeled based on the
bovine rhodopsin crystal structure (PDB code 1U19.pdb)6 using
Modeller 9v1.77 The N-terminus and C-terminus were not included in
any of the models. The final receptor model was energy minimized with
the initially minimized agonist docking pose as described before.
In the final step, the full agonistꢀreceptor complex was embedded in
a pre-equilibrated lipid bilayer consisting of 77 (OPRM) molecules of
1-palmitoyl-2-oleoylphosphatidylcholine (POPC) and solvated with
9215 (OPRM) TIP3P water molecules (box dimensions of 82.8 Å ꢁ
79.4 Å ꢁ 81.5 Å (OPRM)) as described by Urizar et al.78 A short
minimization was applied to the complex embedded in the hydrated
lipid bilayer using AMBER 8 and applying a positional harmonic
S
Supporting Information. Synthesis protocol and char-
b
acterization of 3ꢀ6, amino acid alignments of transmembrane
helices and intra- and extracellular loops, and full IFP bit strings
of 9, 11, 12, and 14. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*For C.d.G: (address) Leiden/Amsterdam Center for Drug
Research, Division of Medicinal Chemistry, Faculty of Science,
VU University Amsterdam, De Boelelaan 1083, 1081 HV Amsterdam,
The Netherlands; (phone) +31 20 5987550; (e-mail) C.de.
Graaf@few.vu.nl. For D.T.: (phone) +32 2 6293295, (e-mail)
datourwe@vub.ac.be.
constraint of 50 kcal/(mol Å) on Cα carbon atoms. The entire system
3
was then subjected to a 500 ps constant pressure molecular dynamics
(MD) simulation. All bonds involving hydrogen atoms were frozen with
the SHAKE algorithm. During the first 250 ps, the Cα carbon atoms
were constrained as previously described and the temperature was
linearly increased from 0 to 300 K. During the last 250 ps, the
temperature was kept constant at 300 K and 1 bar, using a coupling
constant of 0.2 and the Berendsen approach. Interactions were calcu-
lated according to the AMBER 03 force field, using particle mesh Ewald
(MPE) summation to include the long-range electrostratic forces. van
der Waals interactions were calculated using a cutoff of 8.0 Å. The same
upper-bound distance restraints defined for the EL2 Modeller runs were
also used during the MD simulations. Agonist force field parameters
were derived using the Antechamber program,74 and partial charges for
the substrates were derived using the AM1-BCC procedure in Antechamber.
Molecular dynamics snapshots were clustered with the GROMACS g_
cluster tool with respect to TM Cα atoms and according to the
JarvisꢀPatrick method, using a cutoff of 1.0 Å for defining the nearest
neighbor. This yielded four (OPRM) clusters per simulation run. The
cluster representative snapshots were subjected to a short energy
minimization as described above, and the “best” snapshot was chosen
to conduct the subsequent docking studies on. The “best” snapshot was
chosen based on its ability to accommodate 17 in a binding mode in line
with experimental data26,27 by unconstrained docking with Surflex,
version 2.30166 (Figure 4A).
’ ACKNOWLEDGMENT
The authors thank the Fund for Scientific Research Flanders
(FWO-Vlaanderen), the EU Grant Normolife LSHC-CT-2006-
037733, and AstraZeneca for financial support. S.B. is a post-
doctoral fellow of the FWO-Vlaanderen.
’ ABBREVIATIONS USED
Aba, 4-amino-1,2,4,5-tetrahydro-2-benzazepin-3-one; Boc, tert-
butyloxycarbonyl; BSA, bovine serum albumin; Cbz, benzyloxy-
carbonyl; DAMGO, H-Tyr-D-Ala-Gly-NMePhe-Gly-ol; Dmt, 20,60-
dimethyltyrosine; EBAW, 1:1:1:1 ethyl acetate/n-butanol/acetic
acid/water; EL, extracellular loop; FDAA, Nα-(2,4-dinitro-5-
fluorophenyl)-L-alaninamide; GDP, guanosine 50-diphosphate;
GTPγS, guanosine 50-O-(3-thio)triphosphate; GPCR, G-pro-
tein-coupled receptor; GPI, guinea pig ileum; IFP, interaction
fingerprint; IL, intracellular loop; Ile5,6-deltorphin-2, H-Tyr-D-
Ala-Phe-Glu-Ile-Ile-Gly-NH2; i.t., intrathecal; iv, intravenous;
JOM-6, Tyr-c(S-Et-S)[D-Cys-Phe-D-Pen]NH2; MBHA, 4-methyl-
benzhydrylamine; βMePhe, β-methylphenylalanine; OPRD, δ
opioid receptor; OPRK, k opioid receptor; OPRM, μ opioid
receptor; OPRX, nociceptin receptor; Pen, β,β-dimethylcysteine
or penicilamine; ROCS, rapid overlay of chemical structures;
TFA, trifluoroacetic acid; TLC, thin layer chromatography; TM,
transmembrane; TMS, tetramethylsilane
Compounds 9ꢀ14 were docked using standard parameters of Sur-
flex, version 2.301.66 17 was used to generate reference interaction
fingerprints (IFPs) as previously described.68 Seven different interaction
types (negatively charged, positively charged, H-bond acceptor, H-bond
donor, aromatic face-to-edge, aromatic face-to-face, and hydrophobic
interactions) were used to define the IFP. The cavity used for the IFP
analysis consisted of the 26 residues within 4 Å distance from 17 in the
refined receptorꢀligand complex: D3.32, Y3.33, F3.37, G45.46, S45.47,
D45.49, C45.50, T45.51, L45.52, T45.53, F45.54, H45.56, E5.35, K5.39,
F5.43, F5.47, W6.48, I6.51, H6.52, V6.55, K6.58, T6.70, W7.35, H7.36,
I7.39, Y7.43. Standard IFP scoring parameters68 and a Tanimoto coeffi-
cient (Tc-IFP) measuring IFP similarity with the reference 17 pose in
the OPRM receptor model were used to filter (only poses forming an
’ REFERENCES
(1) Kieffer, B. L. Opioid receptors: from genes to mice. J. Pain 2000,
1, 45–50.
(2) Eguchi, M. Recent advances in selective opioid receptor agonists
and antagonists. Med. Res. Rev. 2004, 24, 182–212.
6544
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547

Journal of Medicinal Chemistry
ARTICLE
(3) Lagerstr€om, M. C.; Schi€oth, H. B. Structural diversity of G
protein-coupled receptors and significance for drug discovery. Nat. Rev.
Drug Discovery 2008, 7, 339–357.
(4) Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Motoshima,
H.; Fox, B. A.; Le Trong, I.; Teller, D. C.; Okada, T.; Stenkamp, R. E.;
Yamamoto, M.; Miyano, M. Crystal structure of rhodopsin: a G protein-
coupled receptor. Science 2000, 289, 739–745.
(5) Okada, T.; Ernst, O. P.; Palczewski, K.; Hofmann, K. P. Activa-
tion of rhodopsin: new insights from structural and biochemical studies.
Trends Biochem. Sci. 2001, 26, 318–324.
(6) Li, J.; Edwards, P. C.; Burghammer, M.; Villa, C.; Schertler,
G. F. X. Structure of bovine rhodopsin in a trigonal crystal form. J. Mol.
Biol. 2004, 343, 1409–1438.
(21) Cherezov, V.; Abola, E.; Stevens, R. C. Recent progress in the
structure determination of GPCRs, a membrane protein family with high
potential as pharmaceutical targets. Methods Mol. Biol. 2010, 654,
141–168.
(22) Chien, E. Y.; Liu, W; Zhao, Q; Katritch, V; Han, G. W.; Hanson,
M. A.; Shi, L; Newman, A. H.; Javitch, J. A.; Cherezov, V; Stevens, R. C.
Structure of the human dopamine D3 receptor in complex with a D2/D3
selective antagonist. Science 2010, 330, 1091–1095.
(23) Wu, B.; Chien, E. Y.; Mol, C. D.; Fenalti, G.; Liu, W.; Katritch,
V.; Abagyan, R.; Brooun, A.; Wells, P.; Bi, F. C.; Hamel, D. J.; Kuhn, P.;
Handel, T. M.; Cherezov, V.; Stevens, R. C. Structures of the CXCR4
chemokine GPCR with small-molecule and cyclic peptide antagonists.
Science 2010, 330, 1066–1071.
(7) Rosenbaum, D. M.; Cherezov, V.; Hanson, M. A.; Rasmussen,
S. G. F.; Thian, F. S.; Kobilka, T. S.; Choi, H. J.; Yao, X. J.; Weis, W. I.;
Stevens, R. C.; Kobilka, B. K. GPCR engineering yields high-resolution
structural insights into beta(2)-adrenergic receptor function. Science
2007, 318, 1266–1273.
(8) Cherezov, V.; Rosenbaum, D. M.; Hanson, M. A.; Rasmussen,
S. G. F.; Thian, F. S.; Kobilka, T. S.; Choi, H. J.; Kuhn, P.; Weis, W. I.;
Kobilka, B. K.; Stevens, R. C. High-resolution crystal structure of an
engineered human beta(2)-adrenergic G protein-coupled receptor.
Science 2007, 318, 1258–1265.
(9) Rasmussen, S. G. F.; Choi, H. J.; Rosenbaum, D. M.; Kobilka,
T. S.; Thian, F. S.; Edwards, P. C.; Burghammer, M.; Ratnala, V. R. P.;
Sanishvili, R.; Fischetti, R. F.; Schertler, G. F. X.; Weis, W. I.; Kobilka,
B. K. Crystal structure of the human beta(2) adrenergic G-protein-coupled
receptor. Nature 2007, 450, 383–3U4.
(24) Ballesteros, J. A.; Shi, L.; Javitch, J. A. Structural mimicry in
G-protein-coupled receptors: implications of high-resolution structure
of rhodopsin for structureꢀfunction analysis of rhodopsin-like recep-
tors. Mol. Pharmacol. 2001, 60, 1–19.
(25) de Graaf, C.; Rognan, D. Customizing G protein-coupled
receptor models for structure-based virtual screening. Curr. Pharm.
Des. 2009, 15, 4026–4048.
(26) Fowler, C. B.; Pogozheva, I. D.; Lomize, A. L.; LeVine, H.;
Mosberg, H. I. Complex of an active mu-opioid receptor with a cyclic
peptide agonist modeled from experimental constraints. Biochemistry
2004, 43, 15796–15810.
(27) Fowler, C. B.; Pogozheva, I. D.; LeVine, H.; Mosberg, H. I.
Refinement of a homology model of the mu-opioid receptor using
distance constraints from intrinsic and engineered zinc-binding sites.
Biochemistry 2004, 43, 8700–8710.
(10) Warne, T.; Serrano-Vega, M. J.; Baker, J. G.; Moukhametzianov,
R.; Edwards, P. C.; Henderson, R.; Leslie, A. G. W.; Tate, C. G.; Schertler,
G. F. X. Structure of a beta(1)-adrenergic G-protein-coupled receptor.
Nature 2008, 454, 486–491.
(11) Scheerer, P.; Park, J. H.; Hildebrand, P. W.; Kim, Y. J.; Krauss,
N.; Choe, H. W.; Hofmann, K. P.; Ernst, O. P. Crystal structure of opsin
in its G-protein-interacting conformation. Nature 2008, 455, 497–502.
(12) Jaakola, V. P.; Griffith, M. T.; Hanson, M. A.; Cherezov, V.;
Chien, E. Y. T.; Lane, J. R.; Ijzerman, A. P.; Stevens, R. C. The 2.6 Å
crystal structure of a human A2A adenosine receptor bound to an
antagonist. Science 2008, 322, 1211–1217.
(13) Wang, J. B.; Johnson, P. S.; Persico, A. M.; Hawkins, A. L.;
Griffin, C. A.; Uhl, G. R. Human mu-opiate receptor: cDNA and
genomic clones, pharmacological characterization and chromosomal
assignment. FEBS Lett. 1994, 338, 217–222.
(14) Knapp, R. J.; Malatynska, E.; Fang, L.; Li, X. P.; Babin, E.;
Nguyen, M.; Santoro, G.; Varga, E. V.; Hruby, V. J.; Roeske, W. R.;
Yamamura, H. I. Identification of a human delta-opioid receptor: cloning
and expression. Life Sci. 1994, 54, L463–L469.
(28) Li, J. G.; Chen, C. G.; Yin, J. L.; Rice, K.; Zhang, Y.; Matecka, D.;
de Riel, J. K.; DesJarlais, R. L.; Liu-Chen, L. Y. Asp147 in the third
transmembrane helix of the rat mu opioid receptor forms ion-pairing
with morphine and naltrexone. Life Sci. 1999, 65, 175–185.
(29) Surratt, C. K.; Johnson, P. S.; Moriwaki, A.; Seidleck, B. K.;
Blaschak, C. J.; Wang, J. B.; Uhl, G. R. Mu opiate receptor: charged
transmembrane domain amino-acids are critical for agonist recognition
and intrinsic activity. J. Biol. Chem. 1994, 269, 20548–20553.
(30) Befort, K.; Tabbara, L.; Kling, D.; Maigret, B.; Kieffer, B. L. Role
of aromatic transmembrane residues of the delta-opioid receptor in
ligand recognition. J. Biol. Chem. 1996, 271, 10161–10168.
(31) Befort, K.; Zilliox, C.; Filliol, D.; Yue, S. Y.; Kieffer, B. L.
Constitutive activation of the delta opioid receptor by mutations in trans-
membrane domains III and VII. J. Biol. Chem. 1999, 274, 18574–18581.
(32) Xu, H.; Lu, Y. F.; Partilla, J. S.; Zheng, Q. X.; Wang, J. B.; Brine,
G. A.; Carroll, F. I.; Rice, K. C.; Chen, K. X.; Chi, Z. Q.; Rothman, R. B.
Opioid peptide receptor studies, 11: Involvement of Tyr148, Trp318
and His319 of the rat mu-opioid receptor in binding of mu-selective
ligands. Synapse 1999, 32, 23–28.
(15) Simonin, F.; Befort, K.; Gaveriauxruff, C.; Matthes, H.; Nappey,
V.; Lannes, B.; Micheletti, G.; Kieffer, B. The human delta-opioid
receptor: genomic organization, cDNA cloning, functional expression,
and distribution in human brain. Mol. Pharmacol. 1994, 46, 1015–1021.
(16) Simonin, F.; Gaveriauxruff, C.; Befort, K.; Matthes, H.; Lannes, B.;
Micheletti, G.; Mattei, M. G.; Charron, G.; Bloch, B.; Kieffer, B. Kappa-
opioid receptor in humans: cDNA and genomic cloning, chromosomal
assignment, functional expression, pharmacology, and expression pattern in
the central-nervous-system. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 7006–7010.
(17) Mansson, E.; Bare, L.; Yang, D. M. Isolation of a human kappa-
opioid receptor cDNA from placenta. Biochem. Biophys. Res. Commun.
1994, 202, 1431–1437.
(18) Zhu, J. M.; Chen, C. G.; Xue, J. C.; Kunapuli, S.; Deriel, J. K.;
Liuchen, L. Y. Cloning of a human kappa-opioid receptor from the brain.
Life Sci. 1995, 56, L201–L207.
(19) Pogozheva, I. D.; Przydzial, M. J.; Mosberg, H. I. Homology
modeling of opioid receptorꢀligand complexes using experimental
constraints. AAPS J. 2005, 7, E434–E448.
(33) Spivak, C. E.; Beglan, C. L.; Seidleck, B. K.; Hirshbein, L. D.;
Blaschak, C. J.; Uhl, G. R.; Surratt, C. K. Naloxone activation of mu-
opioid receptors mutated at a histidine residue lining the opioid binding
cavity. Mol. Pharmacol. 1997, 52, 983–992.
(34) Hruby, V. J. Design in topographical space and peptidomimetic
ligands that affect behavior. A chemist’s glipse at the mindꢀbody
problem. Acc. Chem. Res. 2001, 34, 389–397.
(35) Hruby, V. J. Organic chemistry and biology: chemical biology
through the eyes of collaboration. J. Org. Chem. 2009, 74, 9245–9264.
(36) Tourwꢀe, D.; Verschueren, K.; Frycia, A.; Davis, P.; Porreca, F.;
Hruby, V. J.; Toth, G.; Jaspers, H.; Verheyden, P.; VanBinst, G.
Conformational restriction of Tyr and Phe side chains in opioid
peptides: information about preferred and bioactive side-chain topology.
Biopolymers 1996, 38, 1–12.
(37) De Wachter, R.; Brans, L.; Ballet, S.; Van den Eynde, I.;
Urbanczyk-Lipkovska, Z.; Tourwe, D. Influence of ring substitutions
on the conformation and β-turn mimicry of 4-amino-1,2,4,5-tetrahydro-2-
benzazepin-3-one peptide mimetics. Tetrahedron 2009, 65, 2266–2278.
(38) Perola, E.; Charifson, E. Conformational analysis of drug-like
molecules bound to proteins: an extensive study of ligand reorganization
upon binding. J. Med. Chem. 2004, 47, 2499–510.
(20) Chen, Y.; Mestek, A.; Liu, J.; Yu, L. Molecular-cloning of a rat
kappa-opioid receptor reveals sequence similarities to the mu-opioid and
delta-opioid receptors. Biochem. J. 1993, 295, 625–628.
6545
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547

Journal of Medicinal Chemistry
ARTICLE
(39) Montecucchi, P. C.; Henschen, A. Amino-acid-composition
and sequence-analysis of sauvagine, a new active peptide from the skin of
Phyllomedusa sauvagei. Int. J. Pept. Protein Res. 1981, 18, 113–120.
(40) Amiche, M.; Sagan, S.; Mor, A.; Delfour, A.; Nicolas, P.
Characterization of the receptor-binding profile of (H-3) dermorphin
in the rat-brain. Int. J. Pept. Protein Res. 1988, 32, 506–511.
(41) Broccardo, M.; Erspamer, V.; Falconierierspamer, G.; Improta,
G.; Linari, G.; Melchiorri, P.; Montecucchi, P. C. Pharmacological data
on dermorphins, a new class of potent opioid-peptides from amphibian
skin. Br. J. Pharmacol. 1981, 73, 625–631.
(55) Marfey, P. Determination of D-amino acids. 2. Use of a
bifunctional reagent, 1,5-difluoro-2,4-dinitrobenzene. Carlsberg Res.
Commun. 1984, 49, 591–596.
(56) Pꢀeter, J.; Tꢀoth, G.; T€oro€k, G.; Tourwꢀe, D. Separation of
enantiomeric β-methyl amino acids and of β-methyl amino acid contain-
ing peptides. J. Chromatogr., A 1996, 728, 455–465.
(57) Vassiliou, S.; Magriotis, P. A. Improved Sch€ollkopf construction
of quaternary alpha-amino acids: efficient enantioselective synthesis of
integrin LFA-1 antagonist BIRT-377. Tetrahedron: Asymmetry 2006,
17, 1754–1757.
(42) Melchiorri, P.; Negri, L. The dermorphin peptide family. Gen.
Pharmacol. 1996, 27, 1099–1107.
(43) Bryant, S. D.; Jinsmaa, Y.; Salvadori, S.; Okada, Y.; Lazarus,
L. H. Dmt and opioid peptides: a potent alliance. Biopolymers 2003,
71, 86–102.
(58) Vandormael, B.; De Wachter, R.; Martins, J. C.; Hendrickx,
P. M. S.; Keresztes, A.; Ballet, S.; Mallaredy, J. R.; Tꢀoth, F.; Tꢀoth, G.;
Tourwꢀe, D. Asymmetric synthesis and conformational analysis by NMR
spectroscopy and MD of Aba- and α-MeAba-containing dermorphin
analogues. ChemMedChem [Online early access]. DOI: 10.1002/
cmdc.201100314. Published Online: Aug 29, 2011.
(59) Traynor, J. R.; Nahorski, S. R. Modulation by mu-opioid
agonists of guanosine-50-O-(3-[S-35]thio)triphosphate binding to
membranes from human neuroblastoma Sh-Sy5y cells. Mol. Pharmacol.
1995, 47, 848–854.
(60) Darlak, K.; Grzonka, Z.; Krascik, P.; Janicki, P.; Gumulka, S. W.
Structureꢀactivity studies of dermorphin. The role of side chains of
amino acid residues on the biological activity of dermorphin. Peptides
1984, 5, 687–689.
(61) Misicka, A.; Cavagnero, S.; Horvath, R.; Davis, P.; Porreca, F.;
Yamamura, H. I.; Hruby, V. J. Synthesis and biological properties of β-
MePhe3 analogues of deltorphin I and dermenkephalin: influence of
biased χ1 of Phe3 residues on peptide recognition for δ-opioid
receptors. J. Pept. Res. 1997, 50, 48–54.
(62) Kellenberger, E.; Springael, J. Y.; Parmentier, M.; Hachet-Haas, M.;
Galzi, J. L.; Rognan, D. Identification of nonpeptide CCR5 receptor agonists
by structure-based virtual screening. J. Med. Chem. 2007, 50, 1294–1303.
(63) Govaerts, C.; Bondue, A.; Springael, J. Y.; Olivella, M.; Deupi,
X.; Le Poul, E.; Wodak, S. J.; Parmentier, M.; Pardo, L.; Blanpain, C.
Activation of CCR5 by chemokines involves an aromatic cluster between
transmembrane helices 2 and 3. J. Biol. Chem. 2003, 278, 1892–1903.
(64) Govaerts, C.; Blanpain, C.; Deupi, X.; Ballet, S.; Ballesteros,
J. A.; Wodak, S. J.; Vassart, G.; Pardo, L.; Parmentier, M. The TXP motif
in the second transmembrane helix of CCR5. A structural determinant of
chemokine-induced activation. J. Biol. Chem. 2001, 276, 13217–13225.
(65) Valiquette, M.; Vu, H. K.; Yue, S. Y.; Wahlestedt, C.; Walker, P.
Involvement of Trp-284, Val-296, and Val-297 of the human delta-
opioid receptor in binding of delta-selective ligands. J. Biol. Chem. 1996,
271, 18789–18796.
(44) Chandrakumar, N. S.; Yonan, P. K.; Stapelfeld, A.; Savage, M.;
Rorbacher, E.; Contreras, P. C.; Hammond, D. Preparation and opioid
activity of analogs of the analgesic dipeptide 2,6-dimethyl-L-tyrosyl-
N-(3-phenylpropyl)-D-alaninamide. J. Med. Chem. 1992, 35, 223–233.
(45) Schiller, P. W.; Weltrowska, G.; Schmidt, R.; Nguyen, T. M. D.;
Berezowska, I.; Chen, Y.; Lemieux, C.; Chung, N. N.; Wilkes, B. C.;
Carpenter, K. A. Opioid Peptide Analogs with Novel Activity Profiles as
Potential Therapeutic Agents for Use in Analgesia. In Peptide Science:
Present and Future, Proceedings of the 1st International Peptide Sympo-
sium, Kyoto, Japan, Dec 1997; Shimonishi, Y., Ed.; Kluwer Academic
Publishers: Dordrecht, The Netherlands, 1999; pp 665ꢀ669.
(46) Schiller, P. W.; Nguyen, T. M. D.; Berezowska, I.; Dupuis, S.;
Weltrowska, G.; Chung, N. N.; Lemieux, C. Synthesis and in vitro opioid
activity profiles of DALDA analogues. Eur. J. Med. Chem. 2000,
35, 895–901.
(47) Berezowska, I.; Lemieux, C.; Nguyen, T. M. D.; Chung, N. N.;
Schiller, P. W. Opioid Peptide Analogs Containing 20-Hydroxy, 60-
Methyltyrosine in Place of Tyr1 Display Greatly Enhanced δ Agonist
Potency but Unchanged μ Agonist Potency. In Peptides 1998, Proceed-
ings of the 25th European Peptide Symposium, Budapest, Hungary;
Bajusz, S., Hudecz, F., Eds.; Akadꢀemiai Kiadꢀo: Budapest, Hungary, 1998;
pp 718ꢀ719.
(48) Schiller, P. W.; Fundytus, M. E.; Merovitz, L.; Weltrowska, G.;
Nguyen, T. M.-D.; Lemieux, C.; Chung, N. N.; Coderre, T. J. The opioid
μ agonist/δ antagonist DIPP-NH2[ψ] produces a potent analgesic
effect, no physical dependence, and less tolerance than morphine in
rats. J. Med. Chem. 1999, 42, 3520–526.
(49) Tourwꢀe, D.; Mannekens, E.; Nguyen Thi Diem, T.; Verheyden,
P.; Jaspers, H.; Tꢀoth, G.; Pꢀeter, A.; Kertꢀesz, I.; T€or€ok, G.; Chung, N. N.;
Schiller, P. W. Side chain methyl substitution in the delta-opioid
receptor antagonist TIPP has an important effect on the activity profile.
J. Med. Chem. 1998, 41, 5167–5176.
(66) Jain, A. N. Surflex-Dock 2.1: robust performance from ligand
energetic modeling, ring flexibility, and knowledge-based search.
J. Comput.-Aided Mol. Des. 2007, 21, 281–306.
(50) T€omb€oly, Cs.; K€over, K. E.; Pꢀeter, A.; Tourwꢀe, D.; Biyashev, D.;
Benyhe, S.; Borsodi, A.; Al-Khrasani, M.; Ronai, A. Z.; Tꢀoth, G.
Structureꢀactivity study on the Phe side chain arrangement of endo-
morphins using conformationally constrained analogues. J. Med. Chem.
2004, 47, 735–743.
(51) Tourwꢀe, D.; Verschueren, K.; Van Binst, G.; Davis, P.; Porreca, F.;
Hruby, V. J. Dermorphin sequence with high delta-affinity by fixing the Phe
side-chain to trans at χ1. Bioorg. Med. Chem. Lett. 1992, 2, 1305–1308.
(52) Ballet, S.; Frycia, A.; Piron, J.; Chung, N. N.; Schiller, P. W.;
Kosson, P.; Lipkowski, A. W.; Tourwe, D. Synthesis and biological
evaluation of constrained analogues of the opioid peptide H-Tyr-D-Ala-
Phe-Gly-NH2 using the 4-amino-2-benzazepin-3-one scaffold. J. Pept.
Res. 2005, 66, 222–230.
(67) Bonner, G.; Meng, F.; Akil, H. Selectivity of mu-opioid receptor
determined by interfacial residues near third extracellular loop. Eur.
J. Pharmacol. 2000, 403, 37–44.
(68) Marcou, G.; Rognan, D. Optimizing fragment and scaffold
docking by use of molecular interaction fingerprints. J. Chem. Inf. Model.
2007, 47, 195–207.
(69) de Graaf, C.; Foata, N.; Engkvist, O.; Rognan, D. Molecular
modeling of the second extracellular loop of G-protein coupled recep-
tors and its implication on structure-based virtual screening. Proteins
2008, 71, 599–620.
(70) Owens, C. E.; Akil, H. Determinants of ligand selectivity at the
k-receptor based on the structure of the orphanin FQ receptor.
J. Pharmacol. Exp. Ther. 2002, 300, 992–999.
(53) Ballet, S.; Misicka, A.; Kosson, P.; Lemieux, C.; Chung, N. N.;
Schiller, P. W.; Lipkowski, A. W.; Tourwe, D. Bloodꢀbrain barrier
penetration by two dermorphin tetrapeptide analogues: role of lipophi-
licity vs structural flexibility. J. Med. Chem. 2008, 51, 2571–2574.
(54) Kataoka, Y.; Seto, Y; Yamamoto, M. Studies of unusual amino-
acids and their peptides. 6. Synthesis and optical resolutions of beta-
methylphenylalanine and its dipeptide present in bottromycin. Bull.
Chem. Soc. Jpn. 1976, 46, 1081–1084.
(71) ROCS, version 2.3.1; OpenEye Scientific Software: Santa Fe,
NM; www.eyesopen.com.
(72) Boz€u, B.; F€ul€op, F.; Toth, G. K.; Toth, G.; Sz€ucs, M. Synthesis
and opioid binding activity of dermorphin analogues containing cyclic
beta-amino acids. Neuropeptides 1997, 31, 367–372.
(73) Bradford, M. M. A rapid and sensitive method for the quantita-
tion of microgram quantities of protein utilizing the principle of
proteinꢀdye binding. Anal. Biochem. 2008, 72, 248–254.
6546
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547

Journal of Medicinal Chemistry
ARTICLE
(74) Bissantz, C.; Logean, A.; Rognan, D. High-throughput model-
ing of human G-protein coupled receptors: amino acid sequence
alignment, three-dimensional model building, and receptor library
screening. J. Chem. Inf. Comput. Sci. 2004, 44, 1162–1176.
(75) Mosberg, H. I.; Fowler, C. B. Development and validation of
opioid ligandꢀreceptor interaction models: the structural basis of mu vs.
delta selectivity. J. Pept. Res. 2002, 60, 329–335.
(76) Wang, J. M.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case,
D. A. Development and testing of a general AMBER force field.
J. Comput. Chem. 2004, 25, 1157–1174.
(77) Sali, A.; Blundell, T. L. Comparative protein modeling by
satisfaction of spatial restraints. J. Mol. Biol. 1993, 234, 779–815.
(78) Urizar, E.; Claeysen, S.; Deupi, X.; Govaerts, C.; Costagliola, S.;
Vassart, G.; Pardo, L. An activation switch in the rhodopsin family of G
protein-coupled receptors. The thyrotropin receptor. J. Biol. Chem.
2005, 280, 17135–17141.
(79) OMEGA, version 2.2.1; OpenEye Scientific Software: Santa Fe,
NM; www.eyesopen.com.
6547
dx.doi.org/10.1021/jm2003574 |J. Med. Chem. 2011, 54, 6538–6547