Synthesis and biological evaluation of 14-alkoxymorphinans. Part 19: Effect of 14-O-benzylation on the opioid receptor affinity and antagonist potency of naltrexone

Article abstract of DOI:10.1002/hlca.200390187

The 14-O-benzylnaltrexones 3-6 were prepared from naltrexone (2) in several steps. The novel compounds were biologically evaluated in radioligand binding and in [³⁵S]GTPγS functional assays in comparison to the reference compound naltrexone. In the binding assay, compounds 3-6 exhibited preference for κ opioid receptors, while the parent compound naltrexone shows preference for μ receptors. In the functional assay, μ antagonist potency of compounds 3-6 was in the range of naltrexone, while K antagonist potency was considerably higher for most novel compounds in comparison to naltrexone.

Full text of DOI:10.1002/hlca.200390187

Helvetica Chimica Acta Vol. 86 (2003)
2335
Synthesis and Biological Evaluation of 14-Alkoxymorphinans
Part 19
Effect of 14-O-Benzylation on the Opioid Receptor Affinity and Antagonist Potency
of Naltrexone
by Falko Sch¸llnera), Ruth Meditza), Roland Krassniga), G¸nther Morandella), Valery N. Kalininb), Ellen
Sandlera), Mariana Speteaa), Angela Whitec), Helmut Schmidhammer*a), and Ilona P. Berzetei-Gurske*c)
a) Division of Pharmaceutical Chemistry, Department of Pharmacy, University of Innsbruck, Innrain 52a, A-
6020 Innsbruck (Phone (43)512-507-5248; fax (43)512-507-2940; e-mail: helmut.schmidhammer@uibk.ac.at)
b) A.N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, Vavilov Street 28,
117813 Moscow, Russian Federation
c) SRI International, Pharmaceutical Division, 333 Ravenswood Avenue, Menlo Park, CA 94025, USA (Phone
(650)859-3354; fax (650)859-3342; e-mail: ilona.berzetei-gurske@sri.com)
The 14-O-benzylnaltrexones 3 6 were prepared from naltrexone (2) in several steps. The novel compounds
were biologically evaluated in radioligand binding and in [35S]GTPgS functional assays in comparison to the
reference compound naltrexone. In the binding assay, compounds 3 6 exhibited preference for k opioid
receptors, while the parent compound naltrexone shows preference for m receptors. In the functional assay, m
antagonist potency of compounds 3 6 was in the range of naltrexone, while k antagonist potency was
considerably higher for most novel compounds in comparison to naltrexone.
Introduction. Naloxone (1) was the first pure opioid antagonist detected, and it
has become an indispensable tool in opioid research. Both, naloxone and its N-
(cyclopropylmethyl) analogue naltrexone (2) are competitive antagonists at m, k, and d
opioid receptors with some preference for m receptors. The major criterion for the
classification of an agonist effect as being opioid-receptor-mediated is the ability of
these antagonists to competitively antagonize this effect [1][2]. Naloxone (1) is being
used to reverse the potentially lethal respiratory depression caused by neurolept
analgesia or opioid overdose. Among other pharmacological actions, 1 antagonizes the
blood-pressure drop in various forms of shock [3 7], reverses neonatal hypoxic apnea
[1], counteracts chronic idiopathic constipation [8], reduces the food intake in humans
[9][10], and shows beneficial effects in CNS injuries [11].
Naltrexone (2) appears also to be a relatively pure opioid antagonist but with higher
oral efficacy and a longer duration of action than 1 [9]. All these properties make 2
suitable for the management of opioid dependence and provide a new effective
modality for the physician treating addict patients [10][12]. Alcoholism is another
addiction for which treatment with 2 is applied [12].
It was described that 14-O-alkylation (Me, Et) of naloxone (1) and naltrexone (2)
did not significantly alter the in vitro and in vivo potencies of the parent compounds
[13]. We have shown that introduction of a 14-(benzyloxy) group into the opioid
antagonist cyprodime leads to an increase in the antagonist potency at all three opioid

2336
Helvetica Chimica Acta Vol. 86 (2003)
receptors by a factor of five to eight [14][15]. Based on the reported increase in the
antagonist potency of cyprodime, it was of interest to prepare 14-O-benzyl derivatives
of the opioid antagonist 2 and to evaluate these compounds biologically in comparison
to its parent compound. Since benzylation increases lipophilicity, depot preparations
(e.g., transdermal systems) could be developed for therapeutic use.
Results and Discussion. The 14-O-benzylnaltrexone (3) was prepared starting
from 3-O,14-O-dibenzylated ketal 8, which is readily accessible from dihydroxy ketal 7
[14] (Scheme). Compound 8 was hydrolyzed in MeOH/conc. HCl solution to afford
both 3-O,14-O-dibenzylated ketone 13 [14] and 14-O-monobenzylated ketone 3. For
the synthesis of 14-O-(2-methylbenzyl)naltrexone (4), the 3-hydroxy group of ketal 7
was protected with a trityl group ( ! 10) prior to 14-O-alkylation with 2-methylbenzyl
bromide in DMF and NaH as base to give compound 14. Acid hydrolysis afforded
compound 4. The 14-O-(2-chlorobenzyl) and 14-O-(3-chlorobenzyl) derivatives 5 and
6, respectively, were also prepared from ketal 7, which was protected with a benzyl
group to give 3-O-benzylated ketal 9. Compound 9 was then 14-O-alkylated with 2-
chlorobenzyl bromide and 3-chlorobenzyl bromide, respectively, to yield 11 and 12;
acid hydrolysis yielded the desired 6-keto analogues 5 and 6, respectively. The 4-
substituted 14-O-benzyl analogues in another series of morphinans were found to have
less affinity to opioid receptors than their 2- and 3-substituted counterparts
(unpublished data). Therefore, we did not prepare 4-substituted derivatives.
Compounds 3 6 were evaluated in the radioligand binding assay and [35S]GTPgS
assay by means of human opioid receptors transfected into Chinese hamster ovary
(CHO) cells. Receptor-binding studies were conducted employing [3H]DAMGO (m),
[3H]U69,593 (k), and [3H]Cl-DPDPE (d), while in the [35S]GTPgS assay, the
prototypical full agonists DAMGO, U69,593, and DPDPE were used.
In opioid-receptor-binding experiments, compounds 3 6 exhibited preference for k
over m and particularly d opioid receptors, while the parent compound naltrexone (2)

Helvetica Chimica Acta Vol. 86 (2003)
2337
Scheme
shows some preference for m receptors (Table 1). With no exception, the 14-O-benzyl
derivatives of naltrexone exhibited improved k binding affinity, while m and d affinity
was decreased considerably. This is in contrast to earlier findings where 14-O-
benzylation of cyprodime resulted in an increase of affinity to all three opioid-receptor
types [14][15]. In the present study, the 14-O-benzyl derivatives showed particularly
high d/k selectivity ratios ranging from 183 to 636.
Table 1. Opioid-Receptor Binding of Compounds 3 6 and Naltrexone (2)
Ki [nm] Æ s.e.m.
Selectivity ratio
[3H]DAMGO (m)
[3H]U69,593 (k)
[3H]Cl-DPDPE (d)
m/k
d/k
3
4
5
6
0.79 Æ 0.25
3.18 Æ 0.15
3.54 Æ 1.0
0.13 Æ 0.10
0.29 Æ 0.10
0.27 Æ 0.10
0.10 Æ 0.10
0.40 Æ 0.10
23.82 Æ 9.67
140.0 Æ 38.88
171.6 Æ 2.29
27.0 Æ 0.97
6.1
183
483
636
270
22
11.0
13.1
22.9
0.5
2.29 Æ 0.36
0.20 Æ 0.00
Naltrexone (2)
8.70 Æ 0.40
The binding to the d receptor of the 2-Cl-substituted compound 5 was decreased by
one order of magnitude (Ki ˆ 172 nm) compared to the d affinity determined for the 3-
Cl-substituted compound 6 (Ki ˆ 27 nm). However, the binding affinities of compounds
5 and 6 to the k receptor were similar and in the low subnanomolar range, 0.27 and
0.10 nm, respectively. Based on the calculated selectivity ratios, compound 5 was found
to exhibit the highest d/k selectivity ratio of 636, whereas the 3-Cl-substituted
compound 6 showed the largest increase in the m/k selectivity ratio of 23, among the
tested compounds.
Afunctional approach was used to establish the antagonist potency of the new
compounds. In the [35S]GTPgS functional assay, m antagonist potency of compounds
3 6 was in the range of naltrexone (2), while k antagonist potency was considerably
higher for most of the novel compounds in comparison to 2 (Table 2). Interestingly, no

2338
Helvetica Chimica Acta Vol. 86 (2003)
Table 2. Antagonist Ke Valuesof Compounds 3 6 and Naltrexone (2) Determined in the [35S]GTPgS Assay
K2 [nm] Æ s.e.m.
DAMGO (m)
Selectivity ratio
U69,593 (k)
DPDPE (d)
m/k
d/k
3
4
5
6
0.17 Æ 0.05
0.97 Æ 0.26
0.22 Æ 0.05
0.19 Æ 0.02
0.59 Æ 0.04
0.10 Æ 0.03
0.44 Æ 0.39
1.10 Æ 0.13
0.12 Æ 0.06
1.86 Æ 0.16
5.41 Æ 1.73
12.35 Æ 2.36
45.0 Æ 13.03
2.02 Æ 0.61
5.44 Æ 0.75
1.7
2.2
0.2
1.6
0.3
54
28
41
17
Naltrexone (2)
2.9
significant k/m selectivity was noted for the new series of compounds in the functional
assay, while k/d selectivity was reduced in comparison to the findings in the binding
assay.
In conclusion, 14-O-benzylation of naltrexone (2) leads to opioid antagonists with
preference for k opioid receptors, as determined in receptor-binding assay, and higher k
antagonist potency, as established in the functional assay when compared to the parent
compound 2. Since there is evidence that k opioid receptor antagonists have a positive
effect on cocaine addiction, it is feasible that 14-O-benzylnaltrexones may have
therapeutic potential under such conditions.
We wish to thank Prof. Dr. K.-H. Ongania, Institute of Organic Chemistry, University of Innsbruck, for
performing the mass spectra. We further thank the National Institute of Drug Abuse (NIDA), which provided
testing of the compounds under NIDA Contract No. N01DA-7-8072. The work was in part supported by the
Austrian Science Foundation (project P11382-÷MO).
Experimental Part
General. TLC: Polygram SilG/UV254 plates, solvent system CH2Cl2/MeOH/conc. NH4OH soln. 90 :9 :1.
Column chromatography (CC): silica gel 230 400 mesh. M.p.: Kofler melting-point microscope; uncorrected.
1H-NMR Spectra: Varian-Gemini-200 spectrometer; d in ppm rel. to SiMe4 as internal reference, J in Hz. Mass
spectra (CI-MS): Finnigan-MAT-44S apparatus, in m/z. Elemental analyses were performed at the Institute of
Physical Chemistry of the University of Vienna.
14b-(Benzyloxy)-17-(cyclopropylmethyl)-4,5a-epoxy-3-hydroxymorphinan-6-one (3) and 3,14b-Bis(ben-
zyloxy)-17-(cyclopropylmethyl)-4,5a-epoxymorphinan-6-one (13). Asoln. of 8 [14] (4.13 g, 7.3 mmol) in MeOH
(52 ml) and conc. HCl soln. (23 ml) was refluxed for 3 h. After cooling with ice, the soln. was alkalinized with
conc. NH4OH soln. and extracted with CH2Cl2 (3 Â 100 ml). The combined org. phase was washed with H2O,
dried (Na2SO4) and evaporated and the obtained yellow oil (3.64 g) purified by CC (CH2Cl2/MeOH/conc.
NH4OH soln. 250 :2 :0.5): 1.23 g (32%) of pure 13 [14] as a colorless foam and 1.41 g (45%) of 3 as a slightly
yellow oil. Aportion of this material ( 3) was converted into the HCl salt in the usual manner: 3 ¥ HCl. M.p. 2028
1
‡
(dec.; Et2O). H-NMR ((D6)DMSO): 9.61 (s, OH); 8.76 (s, NH ); 7.61 7.29 (m, 5 arom. H); 6.75 (d, J ˆ 8.2,
1 arom. H); 6.68 (d, J ˆ 8.2, 1 arom. H); 5.07 (s, HÀC(5)); 4.80 (d, J ˆ 10.4, 1 H, PhCH2O), 4.25 (d, J ˆ 10.4,
‡
1 H, PhCH2O). CI-MS: 432 ([M ‡ 1] ). Anal. calc. for C27H29NO4 ¥ HCl ¥ 1.3 H2O (491.42): C 65.99, H 6.69,
N 2.85; found: C 65.95, H 6.57, N 2.88.
17-(Cyclopropylmethyl)-4,5a-epoxy-14b-hydroxy-3-(triphenylmethoxy)morphinan-6-one Ethane-1,2-diyl
Acetal (10). Asoln. of trityl bromide (3.8 g, 11.86 mmol), Et 3N (2.41 ml, 15.99 mmol), N,N-dimethylpyridin-
4-amine (DMAP; 14 mg, 0.21 mmol), and 7 (3.0 g, 7.78 mmol) in CH2Cl2 (80 ml) was refluxed for 24 h, cooled,
and washed with H2O (2 Â 100 ml). The org. layer was evaporated, and the obtained yellowish crystals (7.11 g)
were recrystallized from MeOH (15 ml): 4.15 g (85%) of 10. M.p. 203 2058. 1H-NMR (CDCl3): 7.50 7.15
(m, 15 arom. H); 6.60 (d, J ˆ 8.2, 1 arom. H); 6.30 (d, J ˆ 8.2, 1 arom. H); 6.13 (br. s, OH); 4.37 (s, HÀC(5));
‡
4.12 3.80 (m, OCH2CH2O). CI-MS: 628 ([M ‡ 1] ). Anal. calc. for C41H41NO5 ¥ 0.4 MeOH (640.57): C 77.62,
H 6.70, N 2.19; found: C 77.60, H 6.60, N 2.07.

Helvetica Chimica Acta Vol. 86 (2003)
2339
17-(Cyclopropylmethyl)-4,5a-epoxy-14b-[(2-methylbenzyl)oxy]-3-(triphenylmethoxy)morphinan-6-one
Ethane-1,2-diyl Acetal (14). NaH (312 mg, 13.34 mmol; from 520 mg of a 60% NaH dispersion in oil after
washings with petroleum ether) was added to a soln. of 10 (2.0 g, 3.18 mmol) in anh. DMF (20 ml) under N2 at 08
(bath temp.) while stirring. After 20 min, 2-methylbenzyl bromide (0.64 ml, 5.4 mmol) was added, and the
resulting mixture was stirred at r.t. for 20 h. Excess NaH was destroyed carefully by addition of small pieces of
ice. Then, the mixture was diluted with H2O (80 ml) and extracted with CH2Cl2 (1 Â 40 ml, 2 Â 30 ml), the
combined org. phase was washed with H2O (3 Â 100 ml), dried (Na2SO4), and evaporated. The orange oil (1.8 g)
was purified by CC (CH2Cl2/MeOH/conc. NH4OH soln. 250 :2 :0.5): 480 mg of yellowish oil. Crystallization
from MeOH yielded 304 mg (13%) of 14. M.p. 180 1828. 1H-NMR ((D6)DMSO): 7.30 7.15 (m, 19 arom. H);
6.31 (d, J ˆ 8.2, 1 arom. H); 6.23 (d, J ˆ 8.2, 1 arom. H); 4.68 (d, J ˆ 10.2, 1 H, ArCH2O); 4.22 (d, J ˆ 10.2, 1 H,
‡
ArCH2O); 4.17 (s, HÀC(5)); 4.05 3.50 (m, OCH2CH2O); 2.33 (s, MeC6H4). CI-MS: 733 ([M ‡ 1] ). Anal.
calc. for C49H49NO5 ¥ 0.3 MeOH (741.50): C 79.85, H 6.82, N 1.85; found: C 79.94, H 6.85, N 1.96.
17-(Cyclopropylmethyl)-4,5a-epoxy-14b-[(2-methylbenzyl)oxy]-3-hydroxymorphinan-6-one Hydrochlor-
ide (4 ¥ HCl). Asoln. of 14 (300 mg, 0.67 mmol) in MeOH (5 ml) and conc. HCl soln. (0.5 ml) was refluxed
for 10 h. After cooling with ice, the soln. was alkalinized with conc. NH4OH soln. and extracted with CH2Cl2
(1 Â 30 ml, 2 Â 20 ml). The combined org. phase was washed with H2O, dried (Na2SO4), and evaporated and the
obtained yellowish oil (165 mg) purified by CC (CH2Cl2/MeOH/conc. NH4OH soln. 250 :2 :0.5) to yield a
colorless oil (80 mg) which was dissolved in Et2O and treated with Et2O/HCl: 70 mg (23%) of pure 4 ¥ HCl. M.p.
‡
198 2008. 1H-NMR ((D6)DMSO): 9.50 (s, OH); 8.60 (s, NH ), 7.67 7.22 (m, 4 arom. H); 6.74 (d, J ˆ 8.0,
‡
1 arom. H); 6.69 (d, J ˆ 8.0, 1 arom. H); 5.16 (s, HÀC(5)); 2.31 (s, Me). CI-MS: 446 ([M ‡ 1] ). Anal. calc. for
C28H31NO5 ¥ HCl (498.00): C 69.77, H 6.69, N 2.91; found: C 69.53, H 6.76, N 2.56.
3-(Benzyloxy)-17-(cyclopropylmethyl)-4,5a-epoxy-14b-hydroxymorphinan-6-one Ethane-1,2-diyl Acetal
(9). Amixture of 7 (6.9 g, 17.9 mmol), K2CO3 (6.7 g, 48.48 mmol), benzyl bromide (2.34 ml, 19.66 mmol),
and anh. DMF (70 ml) was stirred for 22 h under N2 at r.t. The inorg. material was filtered off and the filtrate
evaporated. The residue was dissolved in CH2Cl2 (80 ml), the soln. washed with H2O (7 Â 50 ml), dried
(Na2SO4), and evaporated. The residue (8.2 g) was recrystallized from MeOH: 7.37 g (87%) of 9. Colorless
crystals. M.p. 130 1318. 1H-NMR (CDCl3): 7.42 7.27 (m, 5 arom. H); 6.75 (d, J ˆ 8.3, 1 arom. H); 6.54 (d, J ˆ
8.3, 1 arom. H); 5.17 (d, J ˆ 11.7, 1 H, PhCH2O); 5.10 (d, J ˆ 11.7, 1 H, PhCH2O); 4.58 (s, HÀC(5)); 4.19 3.73
‡
(m, OCH2CH2O). CI-MS: 476 ([M ‡ 1] ). Anal. calc. for C29H33NO5 (477.58): C 73.24, H 6.99, N 2.95; found:
C 72.91, H 7.09, N 2.94.
3-(Benzyloxy)-14b-[(2-chlorobenzyl)oxy]-17-(cyclopropylmethyl)-4,5a-epoxymorphinan-6-one Ethane-
1,2-diyl Acetal (11). Amixture of 9 (5.82 g, 12.2 mmol), NaH (1.47 g, 61.3 mmol; from 2.45 g of 60% NaH
dispersion in oil after washings with petroleum ether), and anh. DMF (100 ml) was stirred under N2 at 0 58 for
25 min. Then, 2-chlorobenzyl bromide (3.16 ml, 24.3 mmol) was added, and the resulting mixture was stirred for
1 h at 0 58 and then for 3 h at r.t. Excess NaH was destroyed carefully by addition of small pieces of ice. The
mixture was poured into 350 ml of H2O and extracted with CH2Cl2 (1 Â 250 ml, 3 Â 100 ml). The combined org.
phase was washed with H2O (3 Â 100 ml) and brine (2 Â 50 ml), dried (Na2SO4), and evaporated. The obtained
yellow oil (8.30 g) was purified by CC (CH2Cl2/MeOH/conc. NH4OH soln. 250 :4 :0.5): 4.75 g (65%) of pure 11.
1
Colorless oil used as such for the next synthetic step. H-NMR ((D6)DMSO): 7.71 7.31 (m, 9 arom. H); 6.79
(d, J ˆ 8.0, 1 arom. H); 6.57 (d, J ˆ 8.0, 1 arom. H); 5.14 (d, J ˆ 11.6, 1 H, ArCH2O); 5.07 (d, J ˆ 11.6, 1 H,
ArCH2O); 4.81 (d, J ˆ 12, 1 H, ArCH2O); 4.50 (s, HÀC(5)); 4.46 (d, J ˆ 12, 1 H, ArCH2O); 4.01 3.74
(m, OCH2CH2O).
14b-[(2-Chlorobenzyl)oxy]-17-(cyclopropylmethyl)-4,5a-epoxy-3-hydroxymorphinan-6-one Hydrochlor-
ide (5 ¥ HCl). Asoln. of 11 (5.50 g, 9.16 mmol) in MeOH (52 ml) and conc. HCl soln. (23 ml) was refluxed
for 2.5 h. After cooling with ice, the soln. was alkalinized with conc. NH4OH soln. and extracted with CH2Cl2
(1 Â 200 ml, 1 Â 100 ml, 2 Â 50 ml). The combined org. phase was washed with H2O and brine, dried (Na2SO4),
and evaporated. The obtained yellow oil (5.54 g) was purified by CC (CH2Cl2/MeOH/conc. NH4OH soln.
250 :3 :0.5): 2.69 g (63%) of pure 5. Yellow oil. Asmall amount of 5 was dissolved in Et2O and treated with
‡
Et2O/HCl: 5 ¥ HCl. Colorless crystals. M.p. 165 1688. 1H-NMR ((D6)DMSO): 8.63 (br. s, OH); 7.96 (s, NH );
7.50 7.30 (m, 4 arom. H); 6.74 (d, J ˆ 8.0, 1 arom. H); 6.69 (d, J ˆ 8.0, 1 arom. H); 5.19 (s, HÀC(5)); 4.87
‡
(d, J ˆ 11.6, 1 H, ArCH2O); 4.81 (d, J ˆ 11.6, 1 H, ArCH2O). CI-MS: 466 ([M ‡ 1] ). Anal. calc. for
C27H28ClNO4 ¥ HCl ¥ 3.5 H2O (529.44): C 57.35, H 6.42, Cl 12.54, N 2.48; found: C 57.72, H 6.43, Cl 12.37, N 2.09.
3-(Benzyloxy)-14b-[(3-chlorobenzyl)oxy]-17-(cyclopropylmethyl)-4,5a-epoxymorphinan-6-one Ethane-
1,2-diyl Acetal Hydrochloride (12 ¥ HCl). As described for 11, with 9 (0.50 g, 1.1 mmol), NaH (0.14 g,
5.8 mmol), and anh. DMF (15 ml) for 15 min. Then with 3-chlorobenzyl bromide (400 ml, 3 mmol) for 12 h at r.t.
Workup with H2O (350 ml), CH2Cl2 (3 Â 40 ml), H2O (5 Â 40 ml), and brine (1 Â 50 ml). The crude yellow oil

2340
Helvetica Chimica Acta Vol. 86 (2003)
(0.92 g) gave, after CC, 310 mg of a colorless oil that was dissolved in Et2O and treated with Et2O/HCl: 300 mg
‡
(45%) of 12 ¥ HCl. M.p. 124 1258. 1H-NMR ((D6)DMSO): 8.50 (s, NH ); 7.57 7.30 (m, 9 arom. H); 6.92
(d, J ˆ 8.1, 1 arom. H); 6.69 (d, J ˆ 8.1, 1 arom. H); 5.12 (s, PhCH2O); 4.66 4.45 (m, HÀC(5), ArCH2O);
‡
4.00 3.74 (m, OCH2CH2O). CI-MS: 600 ([M ‡ 1] ). Anal. calc. for C36H38ClNO5 ¥ HCl ¥ 4 H2O (708.63):
C 65.33, H 6.37, N 2.12; found: C 65.09, H 6.07, N 1.97.
14b-[(3-Chlorobenzyl)oxy]-17-(cyclopropylmethyl)-4,5a-epoxy-3-hydroxymorphinan-6-one Hydrochlor-
ide (6 ¥ HCl). As described for 5 ¥ HCl, with 12 (500 mg, 0.8 mmol) and MeOH/H2O/conc. HCl soln. 4 :5 :1
(50 ml) for 27 h. Extraction with CH2Cl2 (3 Â 100 ml) and washing with H2O (4 Â 100 ml) and brine (1 Â
100 ml). The crude yellow oil (850 mg) was purified by CC (CH2Cl2/MeOH/conc. NH4OH 250 :2 :0.5): 302 mg
(76%) of 6. Colorless oil. Aportion was dissolved in Et 2O and treated with Et2O/HCl: 6 ¥ HCl. Colorless
‡
crystals. M.p. 147 1508. 1H-NMR ((D6)DMSO): 9.54 (br. s, OH); 8.69 (s, NH ); 7.67 7.37 (m, 4 arom. H); 6.73
(d, J ˆ 8.1, 1 arom. H); 6.67 (d, J ˆ 8.1, 1 arom. H); 5.10 (s, HÀC(5)); 4.78 (s, A rCH2O). CI-MS: 466 ([M ‡
‡
1] ). Anal. calc. for C27H28ClNO4 ¥ HCl ¥ 0.6 Et2O (511.42): C 63.59, H 6.14, Cl 12.77, N 2.52; found: C 63.68,
H 5.95, Cl 12.73, N 2.48.
Receptor-Binding Studies. They were conducted on human opioid receptors transfected into Chinese
hamster ovary (CHO) cells. The m cell lines were maintained in Ham×s F-12 medium supplemented with 10% of
fetal bovine serum and 400 mg/ml of geneticin (G418 sulfate). The d and the k cell lines were maintained in
Dulbecco×s minimal essential medium (DMEM) supplemented with 10% of fetal bovine serum, 400 mg/ml of
geneticin (G418 sulfate), and 0.1% of penicillin/streptomycin. All cell lines were grown to full confluency, then
harvested for membrane preparation. The membranes used for functional assays were prepared in buffer A
(20 mm HEPES, 10 mm MgCl2, and 100 mm NaCl at pH 7.4), and the membranes for binding assays were
prepared in 50 mm Tris buffer, pH 7.7. Cells were harvested by scraping the plates with a rubber policeman and
then centrifuged at 500 Â g for 10 min. The cell pellet was suspended in buffer A or Tris buffer, homogenized in a
Polytron homogenizer, and centrifuged at 20000 Â g for 20 min. The cell pellet was washed in buffer A or Tris,
centrifuged at 20000 Â g for another 20 min, and finally suspended in a small amount of buffer to determine the
protein content. Membranes were placed in small vials at a concentration of 3 mg/ml per vial, stored at À708,
and used as needed.
Routine binding assays were conducted with [3H]DAMGO, [3H]Cl-DPDPE, and [3H]U69,593 to bind to m,
d, and k recetors, resp. For m, d, and k binding, cell membranes were incubated with the appropriate radioligand
and unlabeled drug in a total volume of 200 ml in 96-well plates, usually for 1 h at 258. For routine experiments,
membranes were incubated with the test compounds at concentrations ranging from 10À5 to 10À10 m. After the
incubation, samples were filtered through glass-fiber filters by using a Tomtec cell harvester. Filters were dried
overnight before radioactivity levels were determined. Nonspecific binding was determined by using 1.0 mm of
the unlabeled counterpart of each radioligand.
Full characterization of compounds included analysis of the data for IC50 values and Hill coefficients by
using the program PRISM. Ki Values were calculated by means of the Cheng Prusoff transformation Ki ˆ IC50/
(1 ‡ L/Kd), where L is the radioligand concentration and Kd the binding affinity of the radioligand, as
determined previously by saturation analysis.
[
35S]GTPgS Binding for Functional (Agonist/Antagonist) Determinations. Membranes were prepared as
described above and incubated with [35S]GTPgS (50 pm), GDP (usually 10 mm), and the desired compound, in a
total volume of 200 ml, for 60 min at 258 [16]. Samples were filtered over glass-fiber filters and counted as
described for the binding assays. Adose-response curve with a prototypical full agonist (DAMGO, DPDPE, and
U69593, for m, d, and k receptors, resp.) was conducted in each experiment to identify full and partial agonist
compounds.
High-affinity compounds that demonstrate no agonist activity were tested as antagonists. For each
compound, a full Schild analysis was conducted, utilizing a full agonist-doseÀresponse curve in the presence of
at least three concentrations of the antagonist. The equilibrium dissociation constant (Ke) is calculated from
Ke ˆ a/(DR À 1), where a is the nanomolar concentration of antagonist and DR the virtual shift of the agonist-
concentrationÀresponse curve to the right in the presence of a given concentration of antagonist.
REFERENCES
[1] D. M. Zimmerman, J. D. Leander, J. Med. Chem. 1990, 33, 895.
[2] I. Iijima, J. Minamikawa, A. E. Jacobson, A. Brossi, K. C. Rice, W. A. Klee, J. Med. Chem. 1978, 21, 398.
[3] M. D. Aceto, L. S. Harris, W. L. Dewey, E. L. May, NIDA Res. Monogr. 1979, 27, 330.

Helvetica Chimica Acta Vol. 86 (2003)
2341
[4] N. J. Gurll, D. G. Reynolds, T. Vargish, R. Lechner, J. Pharmacol. Exp. Ther. 1982, 220, 621.
[5] J. W. Holaday, A. I. Faden, Nature (London) 1978, 275, 450.
[6] A. I. Faden, J. W. Holaday, Science (Washington, D.C.) 1979, 205, 317.
[7] A. I. Faden, J. W. Holaday, J. Pharmacol. Exp. Ther. 1980, 212, 441.
[8] M.-J. Kreek, R. A. Schaefer, E. F. Hahn, J. Fishman, Lancet 1983, 261.
[9] W. R. Martin, D. R. Jasinski, P. A. Mansky, Arch. Gen. Psychiat. 1973, 28, 784.
[10] M. S. Gold, C. A. Dackis, A. L. Pottash, H. H. Sternbach, W. J. Annitto, D. Martin, M. P. Dackis, Med. Res.
Rev. 1982, 3, 211.
[11] T. K. McIntosh, A. I. Faden, in −Pharmacological Approaches to the Treatment of Brain and Spinal Cord
Injury×, Eds. G. Stein and B. A. Sabel, Plenum Publishing Corporation, New York, 1988, p. 89.
[12] M.-J. Kreek, in − Handbook of Experimental Pharmacology×, Eds. C. R. Schuler and M. J. Kuhar, Springer,
Berlin, 1996, Vol. 118, p. 563.
[13] R. J. Kobylecki, R. W. Carling, J. A. H. Lord, C. F. C. Smith, A. C. Lane, J. Med. Chem. 1982, 25, 116.
[14] H. Schmidhammer, R. Krassnig, E. Greiner, J. R. Traynor, Heterocycles 1998, 49, 489.
[15] H. Schmidhammer, W. P. Burkard, L. Eggstein-Aeppli, C. F. C. Smith, J. Med. Chem. 1989, 32, 418.
[16] J. R. Traynor, S. R. Nahorski, Mol. Pharmacol. 1995, 47, 848.
Received December 3, 2002