Design and synthesis of novel delta opioid receptor agonists and their pharmacologies

Article abstract of DOI:10.1016/j.bmcl.2009.03.099

We re-examined the accessory site of the 4,5-epoxymorphinan skeleton by camdas conformational analysis in an effort to deign novel δ opioid receptor antagonists. We synthesized three novel compounds (SN-11, 23 and 28) with a 10-methylene bridge and withou

Full text of DOI:10.1016/j.bmcl.2009.03.099

Bioorganic & Medicinal Chemistry Letters 19 (2009) 2792–2795
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of novel delta opioid receptor agonists
and their pharmacologies
a
a
a
c
c
Hiroshi Nagase a, , Yumiko Osa , Toru Nemoto , Hideaki Fujii , Masayuki Imai , Takashi Nakamura ,
*
Toshiyuki Kanemasa c, Akira Kato c, Hiroaki Gouda b, Shuichi Hirono b
a Department of Medicinal Chemistry, School of Pharmacy, Kitasato University, 5-9-1, Shirokane, Minato-ku, Tokyo 108-8641, Japan
b Physical Chemistry for Drug Design, School of Pharmacy, Kitasato University, 5-9-1, Shirokane, Minato-ku, Tokyo 108-8641, Japan
c Pain & Frontier, Discovery Research Laboratories, Shionogi & Co., Ltd, 1405, Gotanda, Koka-cho, Koka, Shiga 520-3423, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
We re-examined the accessory site of the 4,5-epoxymorphinan skeleton by CAMDAS conformational anal-
ysis in an effort to deign novel d opioid receptor antagonists. We synthesized three novel compounds (SN-
11, 23 and 28) with a 10-methylene bridge and without a 4,5-epoxy ring. Among them, compounds SN-
23 (17-isobutyl derivative) and SN-28 (17-methyl derivative) showed very strong agonist activity (over
10 times more than TAN-67). SN-28 also showed high d selectivity. The d agonist activity of SN-23 was
weaker than that of SN-28, but in terms of the d selectivity, SN-23 was higher than that of SN-28. These
unexpected results indicated that the 4,5-epoxy ring, but not the 10-methylene bridge, was an accessory
site in d opioid receptor agonists.
Received 18 February 2009
Revised 14 March 2009
Accepted 23 March 2009
Available online 26 March 2009
Keywords:
Opioid
Naltrexone
Analgesics
d Receptor
Ó 2009 Elsevier Ltd. All rights reserved.
Three types of opioid receptors (
l
, d,
j
) are now well estab-
the postulated two accessory sites would be necessary to be re-
moved to afford full agonist activity. This trial gave us the conclu-
sion which only 4,5-epoxy ring would be an accessory site.
NTI is a potent and selective nonpeptide d antagonist. The ratio-
nale for the design of this antagonist was based on the ‘message-
address’ concept.11 In the binding model of NTI with the d opioid
receptor, we considered the three pharmacophore binding modes
lished not only by pharmacological studies but by molecular bio-
logical studies.1 Recently, nonpeptide d opioid receptor agonists
BW373U86,2 SNC80,3 OMI,4,5 and SIOM6 were described. We have
also reported the highly selective d agonist, TAN-67,7,8 which was
designed from d selective antagonist, Naltrindole (NTI)5,9 by
removing postulated accessory sites,10 that is, the 4,5-epoxy ring
and 10-methylene bridge and converting the indole ring to a quin-
oline ring (Fig. 1).
The key structural features of TAN-67 are a freely rotatable phe-
nol ring and a quinoline nitrogen. The phenol ring of TAN-67 can
rotate to a suitable position by induced fit and the quinoline nitro-
gen can form a hydrogen bond with the d opioid receptor, thereby
inducing agonist activity.7,8 This time we tried to confirm if both of
(ionic,
sage part) and the single delta opioid receptor specific interaction
interaction) at the address part (Fig. 2).
This binding model for NTI led us to the design of the novel
p–p and hydrogen bonds) at the morphinan moiety (mes-
(p–p
selective d receptor agonist, TAN-67. In general, when an agonist
binds a receptor, the receptor changes its structure to accommo-
date the structure of the agonist (induced fit). This change leads
to the next signal transduction, and thus, the agonist produces a
pharmacological effect. On the other hand, since an antagonist
H
OH
H3C
N
N
N
N
H
O
OH
TAN-67
OH
NTI
Figure 1. The structures of TAN-67 and NTI.
* Corresponding author. Tel.: +81 3 5791 6372; fax: +81 3 3442 5707.
E-mail address: nagaseh@pharm.kitasato-u.ac.jp (H. Nagase).
Figure 2. The message part and the address part of NTI.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.099

H. Nagase et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2792–2795
2793
has some extra structural parts (accessory site) that interfere with
the structural change of the receptor, the ligand does not cause
pharmacologic effect even if it binds the receptor. Agonists have
no accessory sites,10 and therefore their binding with the receptor
is sufficient to induce a structural change of the receptor.
On the basis of these ideas, we proposed to remove the possible
accessory sites of NTI to afford a d agonist. Previously, we hypoth-
esized that the accessory sites of NTI were the 4,5-epoxy ring and
the 10-methylene bridge which were removed to give TAN-57.
However, TAN-57 was still an antagonist. We then substituted
the quinoline or quinoxaline rings for the indole ring as the address
site. Both quinoline or quinoxaline rings have a lone pair electron
on the nitrogen, but the indole ring does not. The lone pair electron
would be required to afford agonist activity by additional pharma-
cophore binding with the d receptor. The resulting compound TAN-
67 showed a potent and highly selective d agonist activity, as well
as strong analgesic activity (ED50 = 31.4 mg/kg in the acetic acid
writhing assay).7,8 N-Cyclopropylmethyl (CPM) derivative 1 of
TAN-67 was also d agonist.7
Figure 3 shows the drug design from NTI to TAN-67 and 1. We
postulated that if the 4,5-epoxy ring and the 10-methylene bridge
were removed from NTI, the phenol ring would be able to freely ro-
tate to a suitable position by induced fit and permit the agonistic
interaction with the receptor. On the other hand, compound 2 with
both the 4,5-epoxy ring and the 10-methylene bridge has already
been reported as d antagonist (Fig. 4).12 We questioned if the 4,5-
epoxy ring and the 10-methylene bridge were truly accessory sites.
To clarify this point, we performed conformational analysis of TAN-
67 and model compounds 3 and 4 (Fig. 4).
The conformational analysis of TAN-67 and compounds 3, 4 was
performed using Conformational Analyzer with Molecular Dynam-
ics And Sampling (CAMDAS) 2.1 program.13 The CAMDAS calculations
furnished twelve, four, and two distinct conformers for TAN-67
and compounds 3 and 4, respectively.
Figure 5 shows stereopairs of superimpositions of conformers of
TAN-67, and compounds 3 and 4. The superimpositions correspond
to the best-fit of the quinoline ring. As expected, compound 4 had
the least number of conformers, that is, only two conformers,
which differed slightly in structure around C16 and N17 (Fig. 5C).
On the other hand, both TAN-67 and compound 3 possessed two
kinds of different spatial orientations (O-I and O-II) with respect
Figure 5. The stereopairs of superimpositions of the CAMDAS conformers of TAN-67
(A), and compounds 3 (B), 4 (C).
to the phenol ring (Fig. 5A and B). These two dispositions are re-
lated to the variations of the two torsion angles (T-I and T-II) indi-
cated in Figure 6.
All three compounds can adopt conformers corresponding to O-
I (Fig. 6). On the other hand, conformers corresponding to O-II are
possible only for TAN-67 and compound 3, but not for compound 4.
These results clearly suggest that TAN-67 and compound 3 can un-
dergo significant conformational changes between O-I and O-II, but
compound 4 cannot.
TAN-67 has neither the 4,5-epoxy ring nor the 10-methylene
bridge, and compound 3 has only the 10-methylene bridge. The
range of conformations available to compound 3 is almost the
same as for TAN-67 (Fig. 5A and B). Therefore, we can consider that
the presence of the 10-methylene bridge would not disturb the
structural change of the receptor related to the agonistic effect, be-
cause of the ligand’s ability to undergo conformational change
(Fig. 7A). On the other hand, the presence of the 4,5-epoxy ring
may be responsible for preventing the receptor from undergoing
the structural change due to steric repulsion (Fig. 7B).
These results indicate that only the 4,5-epoxy ring may be an
accessory site, but not the 10-methylene bridge. Taking these
considerations into account, we designed and synthesized a mor-
phinan derivative 3 (SN-28) to obtain a potent and selective d
agonist.
New two compounds with different 17-substituent groups (iso-
butyl, and CPM groups) were also designed and synthesized. The
morphinan structure 5 was synthesized from naltrexone by the re-
ported method.14–16 14-H Morphinan structure 7 was synthesized
from 5 in 3 steps (Scheme. 1),17 and 17-isobutyl derivative 9 was
obtained by the reductive cleavage of the 17-cyclopropylmethyl
substituent in compound 7.18 Compounds 8 and 10 were synthe-
sized from the respective morphinan derivatives 7 and 9 with 2-
aminobenzaldehyde in the presence of a catalytic amount of
CH3SO3H in C2H5OH.7 The resulting compounds 8 and 10 were
demethylated with BBr3 in CH2Cl2 to afford SN-11 and SN-23
respectively (Scheme 1).19 The 17-CPM group of compound 7
was converted to a 17-methyl group by a previously reported
method.20,21 The resulting 17-methyl derivative 12 was converted
to SN-28 (compound 3) as shown in Scheme 2.7,19
H
OH
N
N
N
H
N
Free
rotation
O H
Removal of
accessory site
Accessory
site
OH
TAN-57
Accessory site
OH
NTI
conversion of indole ring
to quinoline ring
TAN-67, 1
Figure 3. The drug design from NTI to TAN-67.
H
R
R2
R1
N
N
N
N
X
OH
OH
2
1
2
3
R = CPM, R = OH, X = -O-:
R = CH3: TAN-67
R = CPM: 1
1
2
R = CH3, R = H, X = -H, -H:
1
2
4
R = CH3, R = H, X = -O-:
Figure 4. The structures of TAN-67 and compounds 2–4.

2794
H. Nagase et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2792–2795
Figure 6. The two-dimensional plots of two selected torsion angles for TAN-67 (A), compounds 3 (B), and 4 (C).
H
H
H3C
RN
N
a)
c)
b)
O
N
7
OCH3
OR
13
R = H: 11
R = CH3:
R = CH3:
R = H: 3 (SN-28)
d)
12
Scheme 2. Reagents and conditions: (a) ClCO2CHClCH3, Cl2CHCHCl2, 100 °C, then
CH3OH, reflux, 80%; (b) 37% HCHO, CH3COONa, 2 M AcOH, rt, 10% Pd–C, H2, rt, 66%;
c) 2-aminobenzaldehyde, CH3SO3H, C2H5OH, 100 °C, 61%; (d) BBr3, CH2Cl2, 0 °C to rt,
48%.
Table 1 shows the binding affinity and selectivity toward opi-
oid
l, d, and j receptors of the three compounds. For purposes
of comparison, NTI and TAN-67 are also presented as standard
compounds. SN-11, 23, and 28 showed high affinities and full
agonist activities (Table 2) for the d receptor. The standard d
Figure 7. The interaction models of three compounds with the receptor. (A) TAN-
67 (red) and compound 3 (blue). (B) compound 4.
antagonist, NTI (l/d = 81.5, j/d = 48.3) and agonist, TAN-67 (l/
d = 197, /d = 509) showed also high selectivity for d receptor
j
in this binding assay. As shown in Table 2, SN-11, 23 and 28 dis-
played better than over 10 times stronger d agonist activities
(EC50 = 0.018, 0.06 and 0.047 nM, respectively) than TAN-67
(EC50 = 0.70 nM) and SNC80 (EC50 = 15.7 nM), which is widely
used as a nonpeptide d agonist. Among these compounds, SN-
11 was the strongest agonist, but its d selectivity was not suffi-
cient. This tendency of CPM derivative was also observed in the
series of TAN-67 derivatives.7 The low selectivity and strong ago-
OH
N
N
a)
O
O
O
O
OCH3
H
OCH3
5
6
H
N
N
d)
b)
c)
O
N
63%
OCH3
OR
8
R = CH3:
R = H: SN-11
7
Table 1
e)
71%
The binding affinity and selectivity of SN-11, 23, and 28 for opioid
l, d and j
receptorsa
H
N
H
d)
N
f)
Compounds
Affinity (Ki, nM)
Selectivity
7
N
O
l
d
j
l/d
j/d
76%
NTI
23.64
284.14
0.77
23.61
11.53
0.29
1.44
0.19
0.19
0.14
14.01
732.45
0.36
4.41
10.2
81.5
197.3
4.1
124.3
82.4
48.3
508.7
1.9
23.2
72.9
OR
10
OCH3
(À)TAN-67
SN-11
SN-23
SN-28
R = CH3:
R = H: SN-23
9
e)
quant.
Scheme 1. Reagents and conditions: (a) SOCl2, pyridine, 0 °C to rt, 95%; (b) PtO2, H2,
CH3OH, rt; (c) 1 M HCl, 50 °C, 65% (2 steps from compound 6); (d) 2-aminobenz-
aldehyde, CH3SO3H, C2H5OH, 100 °C; (e) BBr3, CH2Cl2, 0 °C to rt; (f) 5 M HCl, PtO2,
H2, rt, 73%.
a
Evaluated by ability of each compound to displace [3H]DAMGO (
l
), [3H]DADLE
(d), and [3H]U-69,593 (
guinea pig cerebellum (
j
) binding to membranes of rat cerebellum (
l
and d) or the
j).

H. Nagase et al. / Bioorg. Med. Chem. Lett. 19 (2009) 2792–2795
2795
Table 2
Acknowledgements
d Agonist activities of SN-11, 23 and 28a
Compounds
EC50 (nM)
This work was supported in part by Pola Chemical Industries,
Inc. Ohyama Health Foundation. Inc, and the Tokyo Biochemical
Research Foundation (B1-18). We acknowledge the financial sup-
ports from Shorai Foundation for Science and the Uehara Memorial
Foundation.
(À)TAN-67
(À)TAN-67
SNC80
Met-enk
SN-11
0.7
1.4b
15.7b
1.03
0.018
0.06
0.047
SN-23
SN-28
References and notes
a
Membranes were incubated with [35S] GTP
c
S and GDP with the compound.
1. Dhawan, B. N.; Cesselin, F.; Reisine, T.; Bradley, P. B.; Portoghese, P. S.; Hamon,
M. Pharmacol. Rev. 1996, 48, 567.
2. Chang, K. J.; Rigdon, G. C.; Howard, J. L.; Mcnutt, R. W. J. Pharmacol. Exp. Ther.
1993, 267, 852.
Human recombinant cell membrane (HEK-293) was used in this assay. TAN-67 and
[Met5]-enkephalin were used as the standard d agonists.
b
Ref. 23.
3. (a) Calderon, S.; Rothman, R. B.; Porreca, F.; Flippen-Anderson, R. W.; Mcnutt, R.
W.; Xu, H.; Smith, L. E.; Bilsky, E. J.; Davis, P.; Rice, K. C. J. Med. Chem. 1994, 37,
2125; (b) Bilsky, E. J.; Calderon, S. N.; Wang, T.; Bernstein, R. N.; Davis, P.;
Hruby, V. J.; Mcnutt, R. W.; Rothman, R. B.; Rice, K. C.; Porreca, F. J. Pharmacol.
Exp. Ther. 1995, 273, 359.
nist activity of CPM derivatives may result from the electron-
donating property of CPM group.22
4. Portoghese, P. S.; Sultana, M.; Takemori, A. E. J. Med. Chem. 1990, 33, 1714.
5. Takemori, A. E.; Sultana, M.; Nagase, H.; Portoghese, P. S. Life Sci. 1992, 50, 1491.
6. Portoghese, P. S.; Moe, S. T.; Takemori, A. E. J. Med. Chem. 1993, 36, 2572.
7. Nagase, H.; Kawai, K.; Wakita, H.; Mizusuna, A.; Matsuura, H.; Tajima, C.;
Takezawa, Y.; Endoh, T. Chem. Pharm. Bull. 1998, 46, 1695.
8. Nagase, H.; Yazima, Y.; Fujii, H.; Kawamura, K.; Narita, M.; Kamei, J.; Suzuki, T.
Life Sci. 2001, 68, 2227.
9. Portoghese, P. S.; Sultana, M.; Nagase, H.; Takemori, A. E. J. Med. Chem. 1988, 31,
281.
10. Nogrady, T. In Medicinal Chemistry, A Biochemical Approach; Oxford University
Press: New York, 1985; pp 68–69.
11. Portghese, P. S. Trends Pharmacol. Sci. 1989, 10, 230.
12. Portoghese, P. S.; Nagase, H.; MaloneyHuss, K. E.; Lin, C.-E.; Takemori, A. E.
J. Med. Chem. 1991, 34, 1715.
13. Tsujishita, H.; Hirono, S. J. Comput. Aided Mol. Des. 1997, 11, 305.
14. Sawa, Y. K.; Tada, H. Tetrahedron 1968, 24, 6185.
15. Nemoto, T.; Fujii, H.; Nagase, H. Tetrahedron Lett. 2007, 48, 7413.
16. Nemoto, T.; Fujii, H.; Narita, M.; Miyoshi, K.; Nakamura, A.; Suzuki, T.; Nagase,
H. Bioorg. Med. Chem. 2008, 16, 4304.
17. Osa, Y.; Ida, Y.; Yano, Y.; Furuhata, K.; Nagase, H. Heterocycles 2006, 69, 271.
18. Fujii, H.; Osa, Y.; Ishihara, M.; Hanamura, S.; Nemoto, T.; Nakajima, M.; Hasebe,
K.; Mochizuki, H.; Nagase, H. Bioorg. Med. Chem. Lett. 2008, 18, 4978.
19. Fujii, H.; Narita, M.; Mizoguchi, H.; Murachi, M.; Tanaka, T.; Kawai, K.; Tseng, L.
F.; Nagase, H. Bioorg. Med. Chem. 2004, 12, 4133.
20. Olofson, R. A.; Martz, J. T.; Senet, J.; Piteau, M.; Malfroot, T. J. Org. Chem. 1984,
49, 2081.
21. Nemoto, T.; Fujii, H.; Narita, M.; Miyoshi, K.; Nakamura, A.; Suzuki, T.; Nagase,
H. Bioorg. Med. Chem. Lett. 2008, 16, 4304.
22. We have described that CPM group seemed to have electron-donating property
in the recent reports.a,b (a) Nagase, H.; Yamamoto, N.; Nemoto, T.; Yoza, K.;
Kamiya, K.; Hirono, S.; Momen, S.; Izumimoto, N.; Hasebe, K.; Mochizuki, H.;
Fujii, H. J. Org. Chem. 2008, 73, 8093; (b) Fujii, H.; Imaide, S.; Watanabe, A.;
Nemoto, T.; Nagase, H. Tetrahedron Lett. 2008, 49, 6293.
23. Quock, R. M.; Burkey, T. H.; Varga, E.; Hosohata, Y.; Hosahata, K.; Cowell, S. M.;
Slate, C. A.; Ehlert, F. J.; Roeske, W. R.; Yamamura, H. I. Pharmacol. Rev. 1999, 51,
503.
The more potent activities for d receptor of three compounds,
SN-11, 23, 28 than TAN-67 might be derived from restriction of
rotation of the phenol ring, compared with that of TAN-67. The
phenol ring of TAN-67 may be able to rotate away from the recep-
tor site to form hydrogen bond. On the other hand, the phenol ring
of the three compounds may rotate only within area where could
effectively form hydrogen bond with the receptor (entropy effect).
In summary, to confirm which of the accessory sites of the d
antagonist NTI were required to effect a conformational change
in the receptor, we performed conformational analysis. Based on
the results of these studies, we designed d agonists to confirm
the accessory sites of NTI in comparison with TAN-67. We synthe-
sized three compounds with three types of 17-substituents
(methyl, isobutyl, and CPM groups) and compared them with NTI
and TAN-67 for d selectivity and agonist activity. SN-11, bearing
a 17-CPM group showed the strongest agonist among these three
compounds but the selectivity was not satisfactory. Compounds
SN-23 and 28 with the 10-methylene bridge showed satisfactory
d selectivities and over 10 times stronger agonist activities than
TAN-67, suggesting that the 17-methyl and 17-isobutyl substitu-
ents were almost equally effective for both agonist activity and d
selectivity. From the outcome of these studies, we concluded that
only the 4,5-epoxy ring was an accessory site in NTI, and not the
10-methylene bridge. Currently, we are examining the structure
activity relationships for the above novel derivatives as well as
their in vivo analgesic potency. These results will appear in an
extensive report in the future.