Syntheses and receptor-binding studies of derivatives of the opioid antagonist naltrexone

Article abstract of DOI:10.1016/j.bmc.2003.10.039

Naltrexone (1), which is a member of the group of competitive opioid antagonists, shows a strong affinity for μ-receptors and its derivatives have been notable as novel receptor anatgonists. In this paper, the preparation of several naltrexone derivatives is described; these were used to investigate the role of the oxygenated functional groups in facilitating binding to a series of the opioid receptors. The derivatives showed affinity for opioid μ-receptors which was similar to that of naltrexone, but these compounds, which had masked hydroxyl functional groups, displayed a moderate activity. These results suggest that every oxygenated functional group in naltrexone (1) plays an important role in binding to the opioid receptor.

Full text of DOI:10.1016/j.bmc.2003.10.039

Bioorganic & Medicinal Chemistry 12 (2004) 417–421
Syntheses and receptor-binding studies of derivatives of the opioid
antagonist naltrexone
Koji Uwai,a Hiroko Uchiyama,a Shinobu Sakurada,a Chizuko Kabutob
and Mitsuhiro Takeshitaa,*
aTohoku Pharmaceutical University, 4-4-1 Komatsushima, Aoba-ku, 981-8558 Sendai, Japan
bDepartment of Chemistry, Graduate School of Science, Tohoku University, Aoba-ku, 980-8577 Sendai, Japan
Received 10 September 2003; revised 20 October 2003; accepted 21 October 2003
Abstract—Naltrexone (1), which is a member of the group of competitive opioid antagonists, shows a strong affinity for m-receptors
and its derivatives have been notable as novel receptor anatgonists. In this paper, the preparation of several naltrexone derivatives
is described; these were used to investigate the role of the oxygenated functional groups in facilitating binding to a series of the
opioid receptors. The derivatives showed affinity for opioid m-receptors which was similar to that of naltrexone, but these com-
pounds, which had masked hydroxyl functional groups, displayed a moderate activity. These results suggest that every oxygenated
functional group in naltrexone (1) plays an important role in binding to the opioid receptor.
# 2003 Elsevier Ltd. All rights reserved.
1. Introduction
Although the structure–activity relationships have been
It is generally accepted that there are at least three
major types of homologous opioid receptors (m, d, and
k). A number of putative subtypes of the three major
receptor types have been reported, but definitive evi-
dence for their existence is lacking. The availability of
selective opioid ligands has played an important role in
the cloning and pharmacologic characterization of
opioid receptors and in this regard, a variety of highly
selective ligands have been developed as tools to determine
the effects mediated by these receptors.
extensively investigated,3 the systematic studies regard-
ing the roles of the oxygenated functional groups, espe-
cially for the a-naltrexol (3) [the epimer of b-naltrexol
(2)], have not been reported. This study reports on
structure–activity relationships concerning the oxyge-
nated functional groups of naltrexone (1).
2. Results
2.1. Chemistry
The opioid antagonist naltrexone (1) is useful as a par-
enteral adjunct for treating opioid addiction and alco-
holism. Therefore the structure–activity relationships of
compounds related to naltrexone (1) have been exten-
sively studied and they have shown interesting
characteristics in their actions at opioid receptors.1
Chemical transformation of naltrexone (1) resulted in
various naltrexone derivatives (Scheme 2). Methylation
of naltrexone (1) with CH2N2 in Et2O afforded 3-meth-
oxynaltrexone (4)4 and an unexpected epoxide (5) in
17% and 77% yield, respectively. The structure of 5,
[a]D À144.9ꢀ (c 1.96, CHCl3), was determined based on
the spectral data. The mass spectrum of 5 exhibited a
molecular ion peak [M]+ at 369 which is consistent
b-Naltrexol (2) is reported to be a major metabolite of
naltrexone (1) in mammals2 (Scheme 1). The affinity of
b-naltrexol (2) for the opioid receptors is similar to that
of naltrexone (1).
1
with the molecular formula C22H27O4N. The H and
13C NMR resonances of
5 were unambiguously
assigned by 2D HMQC and HMBC experiments.
NOESY experiments determined that the stereo-
chemistry of the C-6 position in 5 was S. Proof for the
stereochemistry of 5 was provided by X-ray crystal-
lography (Fig. 1).
Keywords: Naltrexone derivatives; Opioid; Receptor binding assay;
Analgesia.
* Corresponding author. Tel.: +81-22-234-4183x2104; fax: +81-22-
275-2013; e-mail: mtake@tohoku-pharm.ac.jp
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.10.039

418
K. Uwai et al. / Bioorg. Med. Chem. 12 (2004) 417–421
It was very interesting that epoxide (5) was obtained in
a stereoselective manner as the conformation of newly
formed epoxide is b. If the reaction mechanism followed
the expected nucleophilic attack of the diazomethane on
the carbonyl group to yield a diazonium betaine, it is
difficult to explain the result, because the a-side of nal-
trexone (1) is sterically hindered compared to b-side.
Our accepted hypothesis of the reaction mechanism is a
concerted cycloaddition that forms a 1,3,4-oxadiazoline
intermediate (Scheme 3).5 The formation of the 1,3,4-
oxadiazoline intermediate is stereochemically deter-
mined (diazomethane access from the b-side). Thus the
stereoselective step of the reaction might occur during
the concerted elimination of N2 from the oxadiazoline.
The carbonyl ylide is not favored due to electrostatic
repulsion between the oxygen atom of the carbonyl
ylide and the oxygen at the 5-position. Thus the internal
rotation of the carbonyl ylide allows formation of a
more stable intermediate, which leads to formation of
the epoxide ring.
3-methyl-a-naltrexol (6)7 in 99% yield. Methylation of
the OH at both C3 and the aliphatic C6 position with
KH gave 3,6-dimethyl-a-naltrexol (7) in 66% yield.
2.2. Opioid receptor binding
Opioid receptor binding data for compounds 1 and 3–7
were obtained with mouse brain or guinea pig cere-
bellum using the procedure of Chang et al.8 Binding was
evaluated by competition between the following selec-
tive radio-ligands: [3H]DAMGO (m) [3H]deltorphin II,
(d) or [3H]U-69,593 (k) and the test compounds chosen.
Table 1 describes the results. Among these compounds,
the Ki values for the known a-naltrexol (3) showed the
Reduction of the C-6 keto group of naltrexone (1) with
NaBH4 gave a-naltrexol (3),6 the epimer of the major
metabolite. Methylation of a-naltrexol (3) was per-
formed with CH3I. Base, in the form of K2CO3 was used
to methylate the aromatic OH at the C3 position, giving
Figure 1. ORTEP drawing of compound (5). Thermal ellipsoids are
drawn at a 30% probability level.
Scheme 1. Metabolic reduction of naltrexone (1) with rat liver S-9
fraction.
Scheme 2. Syntheses of naltrexone derivatives.
Scheme 3. The supposed mechanism of formation of epoxide (5) from naltrexone (1) by diazomethane.

K. Uwai et al. / Bioorg. Med. Chem. 12 (2004) 417–421
Table 1. Receptor binding assays of naltrexone (1) derivatives
419
m
d
k
IC50 (nM)a
Ki (nM)
IC50 (nM)a
Ki (nM)
IC50 (nM)a
Ki (nM)
1
3
4
5
6
7
2.57Æ0.48
2.55Æ0.24
295.0Æ35.5
580.0Æ82.9
>1000
1.55Æ0.29
0.959Æ0.087
110.2Æ13.2
219.1Æ31.3
ND
196.7Æ64.9
213.3Æ6.7
>10000
>10000
>10000
>10000
7.84Æ25.9
8.22Æ5.5
ND
1.78Æ0.36
3.30Æ0.10
4.20Æ70.9
463.3Æ63.8
882.5Æ129.9
463.3Æ63.9
0.711Æ0.144
1.32Æ0.040
167.4Æ28.3
191.9Æ15.0
351.7Æ51.8
184.7Æ25.4
ND
ND
ND
966.7Æ167.1
361.2Æ62.4
DAMGO
Deltorphin II
U-69,593
3.42Æ0.52
1.32Æ0.20
ND
8.33Æ2.37
ND
ND
3.45Æ0.98
ND
ND
ND
9.13Æ100
ND
ND
3.17Æ0.35
ND
ND
ND
ND
a Concentration that gives half-maximal effect. Data are given as the meanÆSEM (n=3).
highest binding affinity to the m-receptors, at the rela-
tively low concentration of ꢁ1.0nM, which is of a
similar order of magnitude to that of the parent com-
pound, naltrexone (1). The remaining ligands showed
moderate activities since their hydroxyl group(s) were
masked with methyl groups or were transformed to the
epoxide; however, all six compounds were m-selective.
grade. Wako gel C-200 (70-150 mm, Wako pure chemi-
cals) was used for column chromatography. Precoated
Kieselgel 60F-254plate (0.25 mm, Merck) was used for
TLC analysis and the spots were detected by the absor-
bance of UV light at 254 nm spraying with Dragen-
dorff’s reagent. The optical rotation was recorded on a
JASCO DIP-360polarimeter. 1H NMR spectra were
measured on JEOL JNM EX270(270MHz), JEOL
JNM 400 (400 MHz), and JEOL JNM 600 (600 MHz)
spectrometers. Chemical shifts (d) are reported as ppm
downfield from tetramethylsilane (TMS), and coupling
constants are given in Hz. Multiplicity is indicated as
follows: s (singlet); d (doublet); dd (doublet of doub-
lets); ddd (doublet of double doublets); t (triplet); m
(multiplet); etc. Mass spectra were recorded on JEOL
JMS DX-303 and JMA-DA 5000 spectrometers.
3. Conclusion
In the present study, naltrexone analogues 3–7 were
synthesized. Their potencies in inhibiting the specific
binding of [3H]DAMGO, a m subunit-selective opioid
agonist, to opioid receptors in the mouse brain, were
evaluated. In addition, their ability to inhibit binding of
[3H]deltorphin II, a d subunit-selective opioid agonist,
to opioid receptors in the mouse brain, and [3H]U-
69,593, a k subunit-selective opioid agonist, to opioid
receptors in the guinea pig cerebellum were assessed.
Our findings demonstrate that the C-3 and C-6 hydro-
xyls or the carbonyl moieties are critical for receptor
binding. Moreover, although there are O-containing
groups (e.g., methoxyl or epoxide) at the C-3 and/or C-6
positions, ‘naked’ O-functional groups (hydroxyl or
ketone) are essential for activity.
4.1.1. (5ꢀ,6ꢀ)-17-(Cyclopropylmethyl)-4,5-epoxymorphi-
nan-3,6,14-triol (6ꢀ-naltrexol (3)). NaBH4 (68.4 mg,
1.80mmol) was added to a solution of naltrexone ( 1)
HCl salt (454 mg, 1.20mmol) in MeOH (12 mL) at
À20 ꢀC. After 30min, acetone was added to the reaction
mixture, which was then concentrated in vacuo. The
residue was diluted with Et2O and then successively
washed with brine, dried over MgSO4 and concentrated
in vacuo. Purification by column chromatography on
silica gel (CHCl3–MeOH, 97:3 v/v) gave the title com-
pound 3 (385 mg, 94%) as a colorless oil: All spectral
data are consistent with previously reported data.6
Despite the fact that the newly synthesized compounds
(5, 6 and 7) did not show the strong activity that we had
anticipated, we expect that some of them could possibly
reveal a potent specific affinity to the putative subtypes
of the three major receptor types (e.g., 3-methoxynal-
trexone (4)). If so, they would be useful tools for the
characterization of new opioid receptors.
4.1.2. (5ꢀ)-17-(Cyclopropylmethyl)-4,5-epoxy-14-
hydroxy-3-methoxymorphinan-6-one (O3-Methyl-(À)-
naltrexone (4)), (5ꢀ)-17-(Cyclopropylmethyl)-4,5-epoxy-
14-hydroxy-3-methoxy-6-methylenemorphinan 6ꢁ-oxide
(5). 0.2 M CH2N2 in Et2O (25 mL) was added to a
solution of naltrexone (1) HCl salt (917 mg, 2.4 mmol)
at 0 ꢀC. After 1 h, the reaction mixture was evaporated.
The residue was dissolved with CHCl3 and successively
washed with a 10% aqueous solution of NH4OH and
brine, dried over Na2SO4 and concentrated in vacuo.
Purification by column chromatography on silica gel
(CHCl3–MeOH, 99:1 v/v and 97:3 v/v) gave O3-Methyl-
(À)-naltrexone (4) (148 mg, 17%) as a colorless oil, and
compound 5 (677 mg, 77%) as colorless needles. Com-
pound 4: All spectral data are consistent with previously
Further studies are currently underway in our laboratory
to identify more effective analogues.
4. Experimental
4.1. Chemistry
All reactions were conducted in oven-dried glassware
under a nitrogen atmosphere. All solvents were solvent

420
K. Uwai et al. / Bioorg. Med. Chem. 12 (2004) 417–421
reported data4. Compound 5: [a]D À144.9ꢀ (c 1.96,
the reaction mixture was dissolved in CHCl3 and suc-
cessively washed with a 1% aqueous solution of HCl, a
saturated solution of aqueous NaHCO3, and brine,
dried over Na2SO4 and concentrated in vacuo. Purifica-
tion by column chromatography on silica gel (CHCl3)
gave the title compound 7 (81.0mg, 68%) as a colorless
1
CHCl3). H NMR (600 MHz, CDCl3) d: 6.72 (d, 1H,
J=8.1 Hz, C-2H), 6.63 (d, 1H, J=8.1 Hz, C-1H), 4.65
(s, 1H, C-5b H), 3.86 (s, 3H, –OCH3), 3.12 (d, 1H,
J=5.5 Hz, C-9H), 3.05 (d, 1H, J=18.0Hz, C-10 b H),
2.91 (d, 1H, J=5.7 Hz, epoxyl OCH2 H), 2,66 (dd, 1H,
J=12.1, 5.1 Hz, C-16H), 2.61 (dd, 1H, J=18.0, 5.5 Hz,
C-10a H), 2.50(dt, 1H, J=2.2, 14.0Hz, C-7 b H), 2.41–
2.22 (m, 3H, C-15H, C-17 2H), 2.28 (d, 1H, J=5.7 Hz,
epoxyl OCH2 H), 2.11 (ddd, 1H, J=12.1, 12.1, 4.0Hz,
C-16H), 1.70(dt, 1H, J=13.2, 3.3 Hz, C-8a H), 1.56-
1.44 (m, 2H, C-8b H, C-15H), 1.13 (dt, 1H, J=14.0, 3.3
Hz, C-7a H), 0.89-0.77 (m, 1H, C-18H), 0.61-0.46 (m,
2H, C-19 2H), 0.19–0.07 (m, 2H, C-19 2H). 13C NMR
(125 MHz, CDCl3) d: 144.22 (s, C-4), 143.04 (s, C-3),
131.77 (s, C-12), 125.39 (s, C-11), 118.93 (d, C-1), 114.13
(d, C-2), 88.66 (d, C-5), 70.23 (s, C-14), 62.33 (d, C-9),
59.71 (s, C-6), 59.71 (t, C-17), 59.12 (q, –OMe), 52.93 (t,
epoxide), 48.68 (s, C-13), 43.89 (t, C-16), 30.81 (t, C-15),
30.38 (t, C-8), 26.71 (t, C-7), 22.57 (t, C-10), 9.40 (d, C-
18), 3.87 (t, C-19), 3.77 (t, C-19). MS m/z: 369 (M+,
100%), 328, 314, 55. High-resolution MS calcd for
C22H27NO4 (M+): 369.1940. Found: 369.1935.
oil: [a]D À158.2ꢀ (c 7.9, CHCl3). H NMR (600 MHz,
1
CDCl3) d: 6.70(d, 1H, J=8.4 Hz, C-2H), 6.55 (d, 1H,
J=8.4 Hz, C-1H), 4.76 (dd, 1H, J=4.4, 1.1 Hz, C-5b
H), 3.85 (s, 3H, C-3–OCH3), 3.81 (dt, 1H, J=11.4, 4.4
Hz, C-6a H), 3.45 (s, 3H, C-6–OCH3), 3.08 (d, 1H,
J=6.2 Hz, C-9H), 3.04 (d, 1H, J=18.7 Hz, C-10b H),
2.66–2.60(m, 1H, C-16H), 2.60(dd, 1H, J=18.7, 7.0
Hz, C-10a H), 2.39–2.32 (m, 1H, C-17H), 2.27–2.20(m,
2H, C-15b H, C-16H), 1.75–1.71 (m, 1H, C-7b H), 1.64
(dt, 1H, J=14.7, 8.1 Hz, C-8b H), 1.58–1.55 (m, 1H, C-
15a H), 1.49–1.45 (m, 1H, C-8a H), 1.24–1.20(m, 1H,
C-7a H), 0.87–0.82 (m, 1H, C-18H), 0.56–0.51 (m, 2H,
C-19 2H), 0.14–0.11 (m, 2H, C-19 2H). 13C NMR
(125 MHz, CDCl3) d: 147.30(s, C-4), 141.66 (s, C-3),
131.15 (s, C-12), 125.86 (s, C-11), 118.07 (d, C-1),
113.84 (d, C-2), 88.39 (d, C-5), 75.72 (d, C-6), 69.93 (s,
C-14), 62.14 (d, C-9), 59.49 (t, C-17), 57.05(q, C-3-
OMe), 56.41 (q, C-6-OMe), 47.20(s, C-13), 43.23 (t, C-
16), 33.58 (t, C-15), 28.66 (t, C-8), 22.73 (t, C-7), 20.30
(t, C-10), 9.34 (d, C-18), 3.91 (t, C-19), 3.72 (t, C-19).
MS m/z: 371 (M+, 100%), 356, 330, 110, 55. High-
resolution MS calcd for C22H29NO4 (M+): 371.2096.
Found: 371.2096.
Crystal data for compound (5): C22H27NO4, Colorless
Plate, Crystal Size=0.40Â0.35Â0.08 mm, a=14.688(5),
b=9.089(3),
c=13.790(5)
A,
b=93.402(5)ꢀ,
V=1837(1)A3, Space group=Monoclinic C2, (No. 4),
Z=4, Dcalc=1.335 g/cm3, T=173 K, Rigaku CCD
diffractometer using Mo-Ka radiation, measured reflec-
tions=8100 (2ꢀ <55ꢀ), independent reflections=4189,
final R=0.0346, Rw=0.036 for 3024 observed reflec-
tions [Io>3s (Io)], GOF=0.98. All calculations were
performed using the TeXsan program (Crystal Structure
Analysis Package, Molecular Structure Corporation,
1985&1999).
4.2. Binding assays
Synaptosomal fractions were prepared from mice spinal
cord and guinea pig cerebellum according to the method
of Cheng et al.8 Briefly, spinal cord or cerebella were
homogenized in a 0.32 M sucrose solution at 0 ꢀC and
centrifuged at 6000g for 15 min. The supernatant was
centrifuged at 40,000g for 30min, and the pellets were
homogenized in 5 mM Tris–HCl buffer (pH 7.4) at 0 ꢀC.
The resulting suspension was centrifuged at 6000g for
15 min. The supernatant was then centrifuged twice at
40000g for 30min each. After the second centrifugation,
the membranes were resuspended in 50mM Tris–HCl
buffer (pH 7.4).
Crystallographic data (excluding structure factors) for
the structure have been deposited with the Cambridge
Crystallographic Data Centre as supplementary pub-
lication nos. CCDC 219210in CIF format. Copies of
the data can be obtained, free of charge, upon request to
CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK,
(fax:44-(0)1223-336033 or e-mail: deposit@ccdc.cam
.ac.uk).
[3H]DAMGO, [3H]deltorphin-II, and [3H]U-69593 were
used to determine the relative affinities for m-, d-, and k-
receptors, respectively. Binding assays were conducted
by incubating an aliquot of the membrane fraction (250
mg for [3H]DAMGO, 350 mg for [3H]deltorphin-II, and
180 mg for [3H]U-69593) containing protease inhibitors
and the labeled ligand (3 nM for [3H]DAMGO or
[3H]deltorphin-II, and 2 nM for [3H]U-69593) in 50mM
Tris–HCl buffer (pH 7.4). After 60min at 25 ꢀC, the
reaction mixture was filtrered through a Whatman GF/
B filter, which was soaked with 0.1% polythyleneimine
and was washed twice with cold Tris–HCl buffer. Filters
were counted in a liquid scintillation counter after
extracting overnight with liquid scintillation fluid (3
mL). Specific binding was determined as the difference
between total binding and that in the presence of excess
(10 mM) unlabeled ligand. Kd and Bmax values were
obtained using Scatchard analysis.
4.1.3.
(5ꢀ,6ꢀ)-17-(Cyclopropylmethyl)-4,5-epoxy-3-
methoxymorphinan-6,14-diol (6). A mixture of a-nal-
trexol (3), (104 mg, 0.31 mmol), MeI (190 mL, 3.05
mmol), K2CO3 (421 mg, 3.05 mmol) in acetone (3 mL)
was refluxed while stirring for 5.5 h. The reaction mix-
ture was then concentrated in vacuo. Purification by
column chromatography on silica gel (CHCl3–MeOH,
49:1 v/v) gave the title compound 6, (108 mg, 99%) as a
colorless oil. All spectral data are consistent with pre-
viously reported data.7
4.1.4. (5ꢀ,6ꢀ)-17-(Cyclopropylmethyl)-4,5-epoxy-3, 6-
dimethoxymorphinan-14-ol (7). A solution of a-nal-
trexol (3) (110mg, .032 mmol) in THF (2mL) was
added, dropwise, to a suspension of KH (54.4 mg, 1.36
mmol) in THF (1 mL) at 0 ꢀC and stirred for 15 min.
Then MeI (100 mL, 1.61 mmol) was added. After 7.5 h,

K. Uwai et al. / Bioorg. Med. Chem. 12 (2004) 417–421
421
References and notes
1997, 412, 35. (b) Larson, D. L.; Jones, R. M.; Hjorth,
S. A.; Schwartz, T. W.; Rhue, W.; Portoghese, P. S. J.
Med. Chem. 2000, 43, 1573. (c) Sakurada, S.; Hayashi,
T.; Yuhki, M.; Fujimura, T.; Murayama, K.; Yone-
zawa, A.; Sakurada, C.; Takeshita, M.; Sato, T.;
Zadina, J. E.; Kastin, A. J.; Sakurada, T. Peptides
2002, 23, 895.
1. (a) De Ponti, F. Curr. Opin. Investig. Drugs 2002, 3, 613.
(b) Ohkawa, S.; Portoghese, P. S. J. Med. Chem. 1998, 41,
4177. (c) Portoghese, P. S.; Sultana, M.; Moe, S. T.;
Takemori, A. E. J. Med. Chem. 1994, 37, 579. (d) Sayre,
L. M.; Larson, D. L.; Takemori, A. E.; Portoghese, P. S.
J. Med. Chem. 1984, 27, 1325. (e) Vizi, E. S.; Foldes, F. F.;
Rich, J.; Knoll, J. Pharmacology 1976, 14, 76.
2. Malspeis, L.; Ludden, T. M.; Bathala, M. S.; Morrison,
B. E.; Feller, D. R.; Reuning, R. H. Res. Commun. Pathol.
Pharmacol. 1976, 14, 393.
3. (a) Dasher, W. E.; Klein, P.; Nelson, W. L. J. Med. Chem.
1992, 35, 2374. (b) Klein, P.; Nelson, W. L. J. Med. Chem.
1992, 35, 4589. (c) Davis, R. D.; Nelson, W. L. J. Med.
Chem. 1995, 38, 570And references are cited therein.
4. (a) Brown, G. P.; Yang, K.; King, M. A.; Rossi, G. C.;
Leventhal, L.; Chang, A.; Pasternak, G. W. FEBS Lett.
5. (a) Shimizu, N.; Bartlett, P. D. J. Am. Chem. Soc. 1978,
100, 4260. (b) Strazzolini, P.; Verardo, G.; Giumanini,
A. G. J. Org. Chem. 1988, 53, 3321.
6. (a) Nelson, T. D.; Davis, R. D.; Nelson, W. L. J. Med.
Chem. 1994, 37, 4270. (b) Chatterjie, N.; Fujimoto, J. M.;
Inturrisi, C. E.; Roerig, S.; Wang, R. I.; Bowen, D. V.;
Field, F. H.; Clarke, D. D. Drug Metab. Dispos. 1974, 2,
401.
7. Simon, C.; Hosztafi, S.; Makleit, S. Tetrahedron 1994, 50,
9757.
8. Chang, K. J.; Cuatrecasas, P. J. Biochem. 1979, 254, 2610.