Synthesis of a novel 6,14-epoxymorphinan derivative and its pharmacology

Article abstract of DOI:10.1016/j.bmcl.2008.10.081

A novel 6,14-epoxymorphinan benzamide derivative (NS22) was synthesized, which showed opioid κ receptor agonistic activity in the [³⁵S]GTPγS binding assay. The antinociceptive effect of NS22 was evaluated in the tail-flick and the hot-plate test. This compound showed a potent antinociceptive activity in mice by sc administration, which was attenuated with nor-BNI (selective opioid κ receptor antagonist).

Full text of DOI:10.1016/j.bmcl.2008.10.081

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6398–6401
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of a novel 6,14-epoxymorphinan derivative and its pharmacology
Toru Nemoto a, Hideaki Fujii a, Minoru Narita b, Kan Miyoshi b, Atsushi Nakamura b, Tsutomu Suzuki b,
Hiroshi Nagase a,
*
a Department of Medicinal Chemistry, School of Pharmacy, Kitasato University, 5-9-1, Shirokane, Minato-ku, Tokyo 108-8641, Japan
b Department of Toxicology, Hoshi University, School of Pharmacy and Pharmaceutical Sciences, 2-4-41, Ebara, Shinagawa-ku, Tokyo 142-8501, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel 6,14-epoxymorphinan benzamide derivative (NS22) was synthesized, which showed opioid
j
receptor agonistic activity in the [35S]GTP
c
S binding assay. The antinociceptive effect of NS22 was eval-
uated in the tail-flick and the hot-plate test. This compound showed a potent antinociceptive activity in
mice by sc administration, which was attenuated with nor-BNI (selective opioid receptor antagonist).
Received 6 September 2008
Revised 16 October 2008
Accepted 17 October 2008
Available online 22 October 2008
j
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Opioid
Naltrexone
Analgesics
6,14-Epoxymorphinan
OH
17
N
15
14
9
8
7
Three types of opioid receptors (l, d, j) are now well-estab-
5
16
10
lished not only by pharmacological studies but also by molecular
13
O
6
O
biological characterizations.1 In naltrexone,
a
l
antagonist
11
1
12
4
(Fig. 1), the 4,5-epoxymorphinan structure (message site) is be-
lieved to influence the intrinsic activity of the ligand with the opi-
oid receptor and the substituents (address site) may help the drug
distinguish among the opioid receptor types.2 Both the morphinan
structure and the 4,5-epoxymorphinan structure may contribute to
three points of association between the drug molecule and the
3
OH
2
naltrexone
Cl
O
O
O
N
Cl
N
CH3
N
CH3
N
receptor site, which include an ionic interaction, a
p–p interaction,
and formation of a hydrogen bond.3,4
U50,488H
U69,593
To obtain potent analgesia without drug dependency, many
highly selective and potent opioid receptor agonists were syn-
j
Figure 1. The structures of naltrexone and representative arylacetamide
j agonists.
thesized, and they are presently available for studies of this recep-
tor. Many selective agonists were arylacetamide derivatives such
as U50,488H and U69,593 in Figure 1,5,6 but all of those derivatives
have serious aversive side effects like psychotomimetic reactions
and were thus excluded from clinical trials.
R
OH
O
CH3
N
14
O
N
N
N
O
N
H
O
O
H
O
We previously reported a novel
j receptor selective agonist,
TRK-820, which has a 4,5-epoxymorphinan skeleton, quite a differ-
ent structure from that of the arylacetamide derivatives (Fig. 2).7,8
Although TRK-820 had no aversive effects, the compound showed
sedative effects and was not approved in clinical trials for post-
operative pain. Eventually TRK-820 was applied as an antipruritics
for patients undergoing dialysis and is now on application for ap-
hydrogen
bond
OH
OH
2
TRK-820
R = OH;
R = H; 1
OH
OH
CH3
N
O
N
N
O
O
O
O
OH
OH
4
3
* 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 structures of TRK-820 and compounds 1–4.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.081

T. Nemoto et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6398–6401
6399
proval. Thus we were able to eliminate aversion effects from a
agonist by using a different basic skeleton.
j
derivative 6 with a stable sulfur ylide (Scheme 1).9 The amide
group of the resulting amide derivative prepared from compound
10 may have similar orientation to that of TRK-820 in Figure 3,
as the oxabicyclo[2.2.1]heptane skeleton fixes the C-ring to more
rigid boat form than that of TRK-820 (Fig. 3).
We then attempted to synthesize other potent
j agonists with a
basic skeleton that differed from TRK-820 to obtain a strong anal-
gesic without the side effects of addiction and sedation. In the
course of the study, we found that TRK-820 derivative 1, without
Herein, we report the synthesis of a 6,14-epoxymorphinan
derivative and its pharmacology.
the 14-hydroxy group, showed very low
j
selectivity (Fig. 2).8 Fur-
thermore, the 6
sic activity nor
a
j
-hydroxyl amide derivative 2 had neither analge-
Compound 5 was synthesized from compound 6 by a previously
reported method.9 The hydroxymethyl group of compound 5 was
treated with CH3SO2Cl in pyridine, and the resulting mesylate 7
selectivity [Ki (nM) = 1.17 ( ), 2.89 ( ), 23.9 (d)],
l
j
in contrast to derivative 3, which lacked the 6-hydroxy group, but
retained selective analgesic activity (ED50 = 0.04 mg/kg, sc, acetic
j
acid writhing tests).8 We postulated that the absence of analgesic
activity for compound 2 resulted from formation of an intramolec-
ular hydrogen bond with 6-hydroxy group (Fig. 2). The amide
group in compound 2 would not participate in a hydrogen bond
with the 14-hydroxy group as the hydrogen atom in the amido
group can only interact with the 6-hydroxy group. To confirm
the effect of 6-hydroxy group, oxazolidinone derivative 4 was syn-
thesized. The Ki value of compound 4 was 4.54 nM (l), 8.66 nM (j),
and 4.82 nM (d), respectively, and was higher than that of com-
pound 2 (no analgesic effect).
Based on the above discussion, we hypothesized that the 14-hy-
droxy group forms a hydrogen bond with the 6-amide side chain
and this orientation of amide side chain assists the adaptation of
the ligand to the
j receptor (Fig. 3). The TRK-820 derivative 1, lack-
ing the 14-hydroxy group, cannot form this hydrogen bond. The
hydrogen bond in TRK-820 may fix the conformation of the C-ring
of the compound into the boat form. The above working hypothesis
led us to design an oxabicyclo[2.2.1]heptane skeleton 5 having the
boat form in the C-ring (Fig. 3).9
We have already reported the synthetic method of the key com-
pound 5 with the novel skeleton by the treatment of morphinan
H3C
H
O
O
N
D ring
O
N
CH2OH
C ring
N
C ring
O
B ring
O
A ring
OH
TRK-820
OCH3
5
Figure 3. The structures of TRK-820 and compound 5. TRK-820 could form
postulated hydrogen bond between 14-OH and 6-N to fix boat conformation.
OH
O
CH2OR
N
N
O
a)
OCH3
OCH3
6
R = H : 5
b)
N
7
:
R = SO2CH3
O
O
CH2R
e)
N
CH2NHCO
c)
Figure 4. Displacement by NS22 of [3H]DAMGO (A), [3H]DPDPE (B) and
[3H]U69,593 (C) binding to membranes of the mouse whole brain or the guinea
pig cerebellum. Values are expressed as % of specific binding of specific radioligand
binding against the log concentration of each compound and values are mean SEM
of 2–4 observations. aIC50 (nM); NS22 0.288, Morphine 1.271, bIC50 (nM); NS22
OCH3
R = N3 : 8
OR
R = CH3 :
R = H
10
: NS22
d)
f)
9
:
R = NH2
Scheme 1. Reagents and conditions: (a) CH3S+O(CH3)2IÀ, NaH, THF, rt, 60%; (b)
CH3SO2Cl, Py, 0 °C; (c) NaN3, DMSO, 110 °C, 89% (2 steps); (d) Pd–C, CSA, MeOH, H2,
rt; (e) (PhCO)2O, Py, rt, 60% (2 steps); (f) C3H7SH, (CH3)3COK, DMF, 150 °C, 61%.
2.073, Morphine >1000, SNC80 ((+)-4-[(a-R)-a-[2S,5R]-4-allyl[2,5-dimethyl-1-pip-
erazinyl]-3-methoxybenzyl]-N,N-diethylbenzamide) 6.411, cIC50 (nM); NS22 0.303,
U50,488 1.136, Morphine 213.1.

6400
T. Nemoto et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6398–6401
Table 1
effects induced by sc administration of NS22 using the hot-plate
and tail-flick tests. A sc administration of NS22 produced an antin-
ociceptive effect and this analgesia was antagonized by the higher
Effect of NS22 on G-protein activation in the mouse whole brain without cerebellum
membranes.a
Treatment
% Stimulation
level of nor-BNI (
b-FNA (
antagonist) or NTI (naltrindole) (d antagonist) (Fig. 6).12
NS22 showed a potent agonistic activity for opioid receptor in
the GTP S binding test and analgesic activity in the mouse hot-
j antagonist) (Fig. 5), but was not antagonized by
l
NS22
NS22 + nor-BNI
NS22 + b-FNA
29.8 2.0
10.7 8.1**
19.2 8.9
j
c
a
Membranes were incubated with [35S]GTP
c
S and GDP with NS22 (10À5 M) in
plate and tail-flick tests. An analgesic effect induced by NS22 was
antagonized by nor-BNI, but not by b-FNA or NTI. These results
indicated that NS22 showed its antinociceptive effect only through
the presence or absence of b-FNA (b-funaltrexamine) (l opioid receptor antagonist,
10À6 M) or nor-BNI (10À6 M). The data are shown as the percentage of basal
[
35S]GTP
c
S binding measured in the presence of GDP and absence of NS22; n = 6.
the
for all three types of opioid receptors. These results suggest that
NS22 may possess opioid receptor agonistic activity, but may
also have antagonistic activity toward and d receptors.
The pronounced analgesic activity of NS22 via the
j opioid receptor, even though NS22 showed binding affinity
**
p < 0.01 versus NS 22 alone.
j
l
was converted to azide 8 with sodium azide in DMSO. The reduc-
tion of compound 8 with Pd–C under H2 afforded amine 9. The
resulting compound 9 was amidated with benzoic anhydride in
pyridine followed by demethylation of amide 10 with 1-C3H7SK
in DMF (Scheme 1).10,11 Finally NS22 was converted to
NS22Áhydrochloride to evaluate the pharmacological activity.11
To evaluate the binding affinities of NS22 for opioid receptors,
we demonstrated the competitive displacement of bound
j
receptor
supports our hypothesis that the orientation of the amide side
chain in NS22, forced by the oxabicyclo[2.2.1]heptane skeleton,
promotes the ligands adaptation to the
j receptor. However,
NS22 still exhibits affinity for
l
and d receptor. We are now
examining the SARs to improve
j
selectivity, and are currently
synthesizing many other 6,14-epoxymorphinan derivatives and
17-cyclopropylmethyl analogs with various 6-amide side chains.
We have also converted the 17-cyclopropylmethyl group to other
substituents and, among them, the 17-benzyl derivatives showed
[3H]DAMGO ([
ligand) (Fig. 4A), [3H]DPDPE (cyclic[
(the d opioid receptor ligand) (Fig. 4B), or [3H]U69,593 (the
oid receptor ligand) (Fig. 4C), using membranes of the mouse
whole brain or the guinea pig cerebellum ( and d: whole brain;
: cerebellum). NS22 showed binding affinity for all three opioid
receptors (Fig. 4).
D
-Ala2,
D
-Leu5]-enkephaline) (the
-Pen2, -Pen5]-enkephalin)
opi-
l opioid receptor
D
D
j
high
j selectivity. These syntheses and the pharmacological data
will be reported in full detail in the near future.
With a more complete investigation of the SARs of the 6,14-
epoxymorphinan compounds, we expect to clarify the most favor-
l
j
able spatial location of the 6-amide side chain for optimum
selectivity. We will further examine the contribution of the boat
form of the C-ring to selectivity. We anticipate that various deriv-
j
We then investigated the ability of NS22 to activate G-proteins
in the mouse whole brain or the guinea pig cerebellum mem-
branes. NS22 showed the GTP
uated with a selective opioid receptor antagonist nor-BNI (nor-
binaltrphimine) (Table 1). We next evaluated the antinociceptive
j
cS binding activity which was atten-
atives of this skeleton will yield new analgesics with limited side
effects of addiction, aversion and sedation.
j
Figure 5. Effect of pre-treatment with nor-BNI (5 or 20 mg/kg, sc) on the NS22 (10 mg/kg, sc)-induced antinociception. Groups of mice were pre-treated with nor-BNI at 24 h,
respectively, before NS22 injection. Each value represents the mean with SEM, n = 6–8. **p < 0.01 or *p < 0.05 versus saline–NS22 group. s SAL (saline) (sc)–SAL, d SAL (sc)–
NS22 (10 mg/kg, sc), N nor-BNI (5 mg/kg, sc)–NS22 (10 mg/kg, sc), ꢀ nor-BNI (20 mg/kg, sc)–NS22 (10 mg/kg, sc).
Figure 6. Effect of pre-treatment with b-FNA (20 mg/kg, sc) or NTI (3 mg/kg, sc) on the NS22 (10 mg/kg, sc)-induced antinociception. Groups of mice were pre-treated with
b-FNA or NTI at 24 h or 30 min before NS 22 injection. Each value represents the mean with SEM of 6–8 trials. s SAL (sc)–SAL, d SAL (sc)–NS22 (10 mg/kg, sc), N b-FNA
(20 mg/kg, sc)–NS22 (10 mg/kg, sc), ꢀ NTI (3 mg/kg, sc)–NS22 (10 mg/kg, sc).

T. Nemoto et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6398–6401
6401
6. Dhawan, B. N.; Cesselin, F.; Raghubir, R.; Reisine, T.; Bradley, P. B.; Portghese, P.
S.; Hamon, M. Pharmacol. Rev. 1996, 48, 567.
7. Nagase, H.; Hayakawa, J.; Kawamura, K.; Kawai, K.; Takezawa, Y.; Matsuura, H.;
Tajima, C.; Endo, T. Chem. Pharm. Bull. 1998, 46, 366.
8. Kawai, K.; Hayakawa, J.; Miyamoto, T.; Imamura, Y.; Yamane, S.; Wakita, H.;
Fujii, H.; Kawamura, K.; Matsuura, H.; Izumimoto, N.; Kobayashi, R.; Endo, T.;
Nagase, H. Bioorg. Med. Chem. 2008, 16, 9188.
9. Nagase, H.; Watanabe, A.; Nemoto, T.; Yamamoto, N.; Osa, Y.; Sato, N.; Yoza, K.;
Kai, T. Tetrahedron Lett. 2007, 48, 2547.
In conclusion, the novel 6,14-epoxymorphinan derivative, NS22
was designed and synthesized to show affinity for all three opioid
receptors and induce its antinociceptive effect only through
oid receptor.
j opi-
Acknowledgments
10. Kawamura, K.; Kawai, K.; Miyamoto, T.; Ooshima, K.; Nagase, H. Heterocycles
1998, 48, 267.
We acknowledge the financial supports from Shorai Foundation
for Science and the Uehara Memorial Foundation. We also
acknowledge the Institute of Instrumental Analysis of Kitasato Uni-
versity, School of Pharmacy for its facilities.
11. N-[(17-Cyclopropylmethyl-6b,14-epoxy-3-hydroxymorphinan-6
a-
yl)methyl]benzamide (NS22): IR (film): 1645 cmÀ1 1H NMR (300 MHz, CDCl3) d:
;
0.04–0.18 (2H, m), 0.42–0.60 (2H, m), 0.93 (1H, m), 1.28 (1H, dt, J = 12.5,
2.0 Hz), 1.37 (1H, m), 1.51–1.90 (5H, m), 2.09 (1H, dt, J = 2.0, 12.5 Hz), 2.23 (1H,
dt, J = 4.5, 12.5 Hz), 2.34 (1H, dd, J = 7.0, 12.58 Hz), 2.52–2.62 (3H, m), 3.10 (1H,
d, J = 18.0 Hz), 3.70 (1H, d, J = 5.5 Hz), 3.85–4.02 (2H, m), 6.61 (1H, d, J = 2.5 Hz),
6.66 (1H, dd, J = 2.5, 8.0 Hz), 6.89 (1H, t, J = 6.0 Hz), 6.93 (1H, d, J = 8.0 Hz),
7.40–7.56 (3H, m), 7.81–7.88 (2H, m), one proton (OH) was not observed. MS
(FAB) m/z = 445 [M+H]+. HRMS (FAB) Calcd for C28H33N2O3 [M+H]+: 445.2491.
Found 445.2504.N-[(17-Cyclopropylmethyl-6b,14-epoxy-3-hydroxymorphinan-
References and notes
1. (a) Knapp, R. J.; Malatynska, E.; Fang, L.; Li, X.; Babin, E.; Nguyen, M.; Santoro,
G.; Varga, E. V.; Hruby, V. J.; Roeske, W. R.; Yamamura, H. I. Life Sci. 1994, 54,
PL463; (b) Wang, J. B.; Johnson, P. S.; Persico, A. M.; Hawkins, A. L.; Griffin, C. A.;
Uhl, G. R. FEBS Lett. 1994, 338, 217; (c) Mansson, E.; Bare, L.; Yang, D. M.
Biochem. Biophys. Res. Commun. 1994, 202, 1431.
6
a-yl)methyl]benzamide hydrochloride (NS22ÁHCl): Mp 181–185 °C (dec). Anal.
Calcd for C28H32N2O3ÁHClÁ0.8H2O: C, 67.88; H, 7.04; N, 5.65. Found: C, 67.85; H,
7.16; N, 5.46.
2. Portghese, P. S. Trends Pharmacol. Sci. 1989, 10, 230.
12. Assessment of antinociception: To prevent tissue damage, we established a 30 s
(hot-plate test) or 10 s (tail-flick test) cut-off time. Each animal served as its
own control, and the latency to response was measured both before and after
drug administration. Antinociception was calculated as percentage of
antinociception according to the following formula: % antinociception = (test
3. Casy, A. F.; Beckett, A. H. J. Pharm. Pharmacol. 1954, 6, 986.
4. Beckett, A. H. J. Pharm. Pharmacol. 1956, 8, 848.
5. (a) Piercey, M. F.; Lahti, R. A.; Schroeder, L. A.; Einspahr, F. J.; Barsuhn, C. Life Sci.
1982, 31, 1197; (b) Lahti, R. A.; Mickelson, M. M.; McCall, J. M.; Von Voigtlander,
P. F. Eur. J. Pharmacol. 1985, 109, 281; (c) Costello, G. F.; Main, B. G.; Barlow, J. J.;
Carroll, J. A.; Shaw, J. S. Eur. J. Pharmacol. 1988, 151, 475; (d) Barber, A.;
Gottschlich, R. Exp. Opin. Invest. Drugs 1997, 6, 1351.
latency À pre-drug
latency)/(cut-off
time À pre-drug
latency) Â 100.
Antinociceptive response represents as the mean SEM of % antinociception.