Probes for narcotic receptor mediated phenomena. Part 28: New opioid antagonists from enantiomeric analogues of 5-(3-hydroxyphenyl)-N-phenylethylmorphan

Article abstract of DOI:10.1016/S0968-0896(02)00219-5

Enantiomeric analogues of 5-(3-hydroxyphenyl)morphan ligands were synthesized and evaluated because of our unexpected finding that opioid antagonists can be obtained in the 5-phenylmorphan series of opioids without sterically hindering the rotation of the phenolic ring. We determined the opioid receptor binding affinity of these new analogues, as well as the efficacy of the more interesting ligands. One of the new compounds [(1R,5S)-(-)-3-[2-(3′- phenylpropyl)-2-azbicyclo[3.3.1]non-5-yl]-phenol, 15] was found to have half of the efficacy of naloxone, a potent opioid antagonist, in the [³⁵ S]GTPγS assay, and two others (1R,5S)-(-)-3-[2-(4′ -- phenylbutyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol, 17, and (1R,5S,1′S)- (+)-3-[2-(1′-methyl-2′-phenylethyl)-2-azabicyclo[3.3.1] non-5-yl]-phenol 26, acted as moderately potent opioid antagonists. X-ray crystallographic structure data were obtained on three compounds. Two of them had three chiral centers; 25 [(1R,5S,1′ R)-(-)-3-[2-(1′-methyl-2′-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol] was determined to have the 1R,5,S,1′R configuration, and 26 the 1R,5S,1′S configuration. Since (1S,5R)-(+)-2-bromo-5-[2-(2′-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (32) was a position isomer of (1S,5R)-(+)-4-bromo-3-[ 2-(2′ -phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (30), and both showed the same ¹ H NMR spectrum, the structure of 32 was unequivocally determined by X-ray structure analysis. Copyright

Full text of DOI:10.1016/S0968-0896(02)00219-5

Bioorganic & Medicinal Chemistry 10 (2002) 3319–3329
Probes for Narcotic Receptor Mediated Phenomena. Part 28:
New Opioid Antagonists from Enantiomeric Analogues of
5-(3-Hydroxyphenyl)-N-phenylethylmorphan
Akihiro Hashimoto,a Arthur E. Jacobson,a Richard B. Rothman,b Christina M. Dersch,b
Clifford George,c Judith L. Flippen-Andersonc and Kenner C. Ricea,*
aLaboratory of Medicinal Chemistry, Building 8, Room B1-23, National Institute of Diabetes and Digestive and Kidney Diseases,
National Institutes of Health, Bethesda, MD20892-0815, USA
bClinical Psychopharmacology Section, National Institute on Drug Abuse, Addiction Research Center, Baltimore, MD 21224, USA
cLaboratory for the Structure of Matter, Naval Research Laboratory, Washington, DC 20375, USA
Received 25 January 2002; accepted 16 April 2002
Abstract—Enantiomeric analogues of 5-(3-hydroxyphenyl)morphan ligands were synthesized and evaluated because of our unex-
pected finding that opioid antagonists can be obtained in the 5-phenylmorphan series of opioids without sterically hindering the
rotation of the phenolic ring. We determined the opioid receptor binding affinity of these new analogues, as well as the efficacy of
the more interesting ligands. One of the new compounds [(1R,5S)-(ꢀ)-3-[2-(30-phenylpropyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol,
15] was found to have half of the efficacy of naloxone, a potent opioid antagonist, in the [35S]GTPgS assay, and two others
(1R,5S)-(ꢀ)-3-[2-(40-phenylbutyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol, 17, and (1R,5S,10S)-(+)-3-[2-(10-methyl-20-phenylethyl)-2-
azabicyclo[3.3.1]non-5-yl]-phenol, 26, acted as moderately potent opioid antagonists. X-ray crystallographic structure data were
obtained on three compounds. Two of them had three chiral centers; 25 [(1R,5S,10R)-(ꢀ)-3-[2-(10-methyl-20-phenylethyl)-2-azabi-
cyclo[3.3.1]non-5-yl]-phenol] was determined to have the 1R,5S,10R configuration, and 26 the 1R,5S,10S configuration. Since
(1S,5R)-(+)-2-bromo-5-[2-(20-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (32) was a position isomer of (1S,5R)-(+)-4-bromo-
3-[2-(20-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (30), and both showed the same H NMR spectrum, the structure of 32
1
was unequivocally determined by X-ray structure analysis.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
found to be as potent as morphine, levorphanol, and
metazocine as an antinociceptive.5 Only agonists were
May and his collaborators began to explore the effect of
simplification of the structure of the 5-ring morphine-
like compounds (4,5-epoxymorphinans) in the 1950’s.1ꢀ4
They hypothesized that molecular simplification might
reduce undesired pharmacological effects (tolerance,
physical dependence, abuse liability, respiratory depres-
sion, and constipation, were well known at that time),
while retaining the analgesic activity of the parent epox-
ymorphinan. Among the various types of compounds
that they synthesized were the m-substituted phenylcy-
clohexanes, bicyclic compounds that became known as
the 5-phenylmorphans.5 The parent compound in this
series, [(Æ)-5-m-hydroxyphenyl-N-methylmorphan ((Æ)-
3-(2-methyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol, 1] was
found in this series until May et al. separated the enan-
tiomers of 1. The (1R,5S)-(ꢀ)-enantiomer (2) was about
4 times more potent than morphine in the mouse hot
plate assay,6 and the (1S,5R)-(+)-enantiomer (3) was an
agonist-antagonist. It was found to have weak (nalor-
phine-like) antagonist activity and it was morphine-like
in its antinociceptive potency.6,7 Ong et al. noted6 that
potent antagonists could not be prepared from analo-
gous potent phenylmorphan agonists. Modification of
the N-substituent in the phenyl-axial 6,7-benzomor-
phan, 4,5-epoxymorphinan, and morphinan series (e.g.,
by replacing the N-methyl with an N-cyclopro-
pylmethyl, N-propyl, or N-allyl moiety) was known to
convert potent opioid agonists to potent opioid antago-
nists. This did not occur in the 5-phenylmorphan series.
Ong et al. theorized6 that it might be necessary for the
phenylmorphans to have an additional fused ring that
*Corresponding author. Tel.: +1-301-496-1856; fax: +301-402-0589;
e-mail: kr21f@nih.gov
0968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(02)00219-5

3320
A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
Chart 1.
would fix the position of the phenolic ring, in order to
obtain opioid antagonist activity. That idea could con-
ceivably be extended to say that the agonist and antago-
nist behavior of opioids might be dependent on the
position of the phenyl ring in three-dimensional space.
antagonist was not discovered until quite recently.
Thomas et al. found8 that they could obtain potent 5-
phenylmorphan-based antagonists using a method initi-
ated by Zimmerman et al.9,10 with the structurally
reminiscent phenylpiperidines. Prior to Zimmerman’s
work, none of the phenylpiperidine series of opioid
agonists had been converted to opioid antagonists.
A ‘pure’ high affinity potent phenylmorphan opioid
Scheme 1. Reagents and conditions: (a) Ac2O, 90 ꢁC, 1 h; (b) 2,2,2-trichloroethoxycarbonyl chloride, NaHCO3, toluene, reflux, 15 h; (c) Zn, AcOH/
H2O, rt, 3 h; (d) NaOH, MeOH, 0 ꢁC.

A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
3321
Zimmerman et al. obtained antagonists in the 4-phe-
nylpiperidine series through steric hindrance of the
rotation of the aromatic ring, using alkyl groups on the
piperidine ring that were vicinal to the phenyl ring.
Using that analogy, Thomas et al.8 added an alkyl
moiety to a phenylmorphan which could sterically hin-
der the rotation of the phenylmorphan’s phenolic ring,
and found that the resulting racemic 9b-methyl-5-(3-
hydroxyphenyl)-N-phenylethylmorphan was a pure,
potent, modestly selective m-opioid antagonist (as
determined by [35S]GTPgS assays).8 Steric hindrance
was implicated because racemic 5-(3-hydroxyphenyl)-N-
phenylethylmorphan (14), the comparable compound
with a freely rotating phenolic ring (without a 9b-methyl
group), was known to be a weak agonist.11 Thus, the
data of Thomas et al.,8 like those of Ong et al., impli-
cated the spatial position of the phenyl ring in the in
vivo behavior of the 5-phenylmorphans. Modification
of that spatial positioning, then, should modify the in
vivo behavior of the compound. Indeed,Awaya et al.12
have noted that (+)-9a-methyl-5-(3-hydroxyphenyl)-N-
phenylethylmorphan had weak antagonist actions
(about half as potent as nalorphine), and its (ꢀ)-enan-
tiomer had a weak agonist effect (about one-tenth as
potent as morphine in the hot plate assay). It is likely
that the spatial position of the phenyl ring is somewhat
different in a 9a-methyl and a 9b-methyl substituted
phenylmorphan. Unfortunately, since the N-phenylethyl
Scheme 2. Reagents and conditions: (a) appropriate alkyl halide, DMF, NaHCO3, 80 ꢁC, 2 h; (b) NaOH, MeOH.
Scheme 3. Reagents and conditions: (a) 1-phenylpropane-2-one, TsOH, benzene, reflux, 15 h; (b) NaBH3CN, MeOH, AcOH; (c) HBr; (d) recryst., MeOH.

3322
A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
Figure 1. X-ray crystallographic structures of (1R,5S,10R)-(ꢀ)-3-[2-(10-methyl-20-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (25), (1R,5S,10S)-
(+)-3-[2-(10-methyl-20-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (26), and (1S,5R)-(+)-2-bromo-5-[2-(20-phenylethyl)-2-azabicyclo[3.3.1]non-
5-yl]-phenol (32). The figure is drawn using the experimentally determined coordinates with thermal parameters at 20% probability levels.
analogue of the 9a-methyl compound was not prepared,
and since Thomas et al.8 noted that their racemic
N-methyl analogue in the 9b-methyl series was a low
affinity agonist [very reminiscent of the weak in vivo
effect of the (+)-racemic N-methyl analogue in the 9a-
methyl series], it is difficult to conclude that the spatial
position of the phenyl ring in the 9b-methyl N-pheny-
lethyl series is best for antagonist activity.
essential for antagonist activity, we were surprised to find
that the angular relationship between the aromatic and
piperidine rings in the 5-phenylmorphans was, in fact,
not essential. We recently reported the preparation and
in vitro activity of the enantiomers of 5-(3-hydroxy-
phenyl)-N-phenylethylmorphan,13 in which the phenolic
ring can freely rotate. We hypothesized that one of the
enantiomers of the racemate described, by May et al.,11
as a weak agonist might have appreciable antagonist
activity; the racemate was not examined for antagonist
activity. We found that the 1R,5S-(ꢀ)-enantiomer (12)
Since it seemed clear that a relatively fixed spatial posi-
tion of the phenyl ring in the 5-phenymorphans was

A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
3323
Scheme 4. Reagents and conditions: (a) Br2, CS2, 0 ꢁC; (b) HBr, DMSO, rt, 24 h; (c) Br2, CS2; (d) (R)-(ꢀ)- or (S)-(+)-2-phenylpropionic acid,
.
DPPA, Et3N, DMF(e) BH 3 THF, THF, reflux; (f) HCl, MeOH.
was indeed a pure, albeit lower affinity, opioid antago-
nist similar to the comparable racemic 9b-methyl-5-(3-
hydroxyphenyl)-N-phenylethylmorphan of Thomas et
al.,8 using the same [35S]GTPgS assay criteria. In order
to expand upon that unanticipated finding, we have
now synthesized and determined the opioid binding
affinities of a series of enantiomeric analogues of 5-(3-
hydroxyphenyl)-N-phenylethylmorphan (Chart 1), and
we have examined the efficacy of the more interesting
analogues to see whether even more potent antagonists
can be found among these sterically unhindered
phenylmorphan enantiomers.
nate in dimethylformamide, followed by deacetylation in
aqueous sodium hydroxide, gave enantiomers 12, 13 and
15 through 24 (Scheme 2). The yield was improved if 8 or
9 was initially deacetylated to 10 or 11, and then N-alky-
lated with the bromide to give the desired compounds. N-
Alkylation with 1-phenylpropane-2-one was accom-
plished using, p-toluenesulfonic acid in toluene, followed
by reduction with sodium cyanoborohydride to give 25–
28 (Scheme 3). The molecular structures of 25 and 26,
with three chiral centers, were unequivocally determined
by X-ray crystallography. Thus, 25 was found to be the
1R,5S,10R analogue (Fig. 1), and 26 (Fig. 1) was the
1R,5S,10S analogue. Bromination of 12 or 13 with a
slight excess of bromine in CS2 gave 29 or 30, respec-
tively (Scheme 4). These were reacted with HBr in
dimethylsulfoxide16 to give 31 and 32 (Scheme 4). Since
both of the position-isomers 32 and 30 showed the same
1H NMR spectrum, the structure of 32 was determined
by X-ray structural analysis (Fig. 1). Compound 32 was
found to be (1S,5R)-(+)-2-bromo-5-[2-(20-phenylethyl)-
2-azabicyclo[3.3.1]non-5-yl]-phenol. Bromination of (Æ)-
3-[2-(20-phenylethyl)-2-azabicyclo [3.3.1]non-5-yl]-phenol
(14) gave 33 (Scheme 4). Coupling of 10 with the enan-
tiomers of 2-phenylpropionic acid using diphenylpho-
sphoryl azide followed by treatment with BH3-THF
complex, gave 34 and 35 (Scheme 4).
Chemistry
Optically pure 2 and 3 were prepared according to the lit-
erature.6,14,15 Acetic anhydride acetylation of 2 or 3 at
90ꢁC gave the enantiomers of 4 or 5, respectively (Scheme
1). Without an acetyl protective group, it was difficult to
obtain the desired compounds. Demethylation of 4 or 5
was accomplished using 2,2,2-trichloroethyl chlor-
oformate, to give the intermediate 6 or 7, respectively.
Carbamates 6 or 7 were converted to the secondary
amine 8 or 9 using zinc-acetic acid. Reaction of 8 or 9 with
the corresponding alkyl bromide using sodium bicarbo-

3324
A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
Table 1. Opioid receptor binding affinity of synthesized phenylmor-
phans18,19
Table 2. Inhibition of [35S]GTPgS binding in guinea pig caudate sti-
mulated by DAMGO (m, SNC80 (d), and U69,593 (k selective opioid
agonists20
Compd
Ki (nMÆSD)
d (SNC80)b
Compd
Ki (nMÆSD)
d (SNC80)b
m (DAMGO)a
k (U69,593)c
m (DAMGO)a
k (U69,593)c
12
13
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
14.4Æ0.7d
27Æ1.7d
42Æ2
180Æ20d
240Æ10d
700Æ50
180Æ10d
90Æ7d
12
13
15
17
21
25
7.5Æ0.8d
5.5Æ0.2d
3.2Æ0.2
6.5Æ0.4
120Æ10
50Æ5
223Æ15d
10Æ0.6d
28Æ3
28Æ1.2d
16Æ0.8d
17Æ2
390Æ30
380Æ10
360Æ30
1300Æ90
1300Æ80
2000Æ140
1100Æ60
760Æ60
70Æ10
100Æ4
1000Æ40
700Æ50
30Æ2
30Æ3
72Æ5
230Æ10
220Æ10
540Æ10
30Æ2
1200Æ130
600Æ70
>1000
490Æ50
50Æ5
420Æ30
100Æ9
22Æ3
2000Æ300
550Æ40
26
Naloxone
9.3Æ1
1.4Æ0.1e
25Æ2.0e
11Æ1.2e
80Æ5
900Æ130
660Æ60
aDAMGO (d-Ala2,MePhe4Gly-ol5)enkephalin), agonist selective for
m-opioid receptor.
150Æ10
200Æ10
29Æ2.8
31Æ2.3
350Æ30
340Æ40
1100Æ60
1700Æ70
900Æ60
1400Æ70
>6000
1000Æ70
290Æ40
380Æ30
240Æ10
50Æ4
bSNC80 ([(+)-4-[(aR)-a-(2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)-3-
methoxylbenzyl]-N,N0-diethylbenzamide), agonist selective for d-
opioid receptor.
170Æ20
1400Æ140
1100Æ130
1800Æ200
7000Æ700
1600Æ120
>4300
90Æ8
cU69,593 (trans-3,4-dichloro-N-methyl[2-(1-pyrrolidinyl)cyclohexyl]
benzeneacetamide), agonist selective for k opioid receptor.
dData from ref 13.
190Æ30
7700Æ630
11000Æ800
7000Æ700
11500Æ700
>6400
eData from ref 21.
4400Æ600
2100Æ100
2400Æ100
840Æ60
560Æ30
920Æ50
1400Æ70
1R,5S series (26), had essentially the same affinity for
both m and k (k/m =1.5) opioid receptors. The 20S
hydroxyl compound in the 1R,5S series, compound 21,
has greater k/m selectivity (k/m =37) than the formerly
reported 2 (k/m =13), and the 20R compound 22 showed
less m- (Ki=80 nM) and d- (Ki=900 nM) receptor affi-
nity and had less k/m selectivity (k/m<10) than 21. All of
the 1R,5S compounds were found to have higher m
opioid receptor affinity than their enantiomeric 1S,5R
relatives. The introduction of bromine into the phenolic
ring of the racemic N-phenylethyl-5-phenylmorphan,
ortho and para to the phenolic hydroxyl (33), was found
to greatly reduce affinity for any of the opioid receptors.
Conceivably, the bromine’s electron withdrawing effect
on the aromatic ring perturbed the necessary hydrogen-
bonding interaction of the phenolic hydroxyl group
with specific amino acids in the opioid receptors.
aDAMGO (d-Ala2, MePhe4Gly-ol5)enkephalin), agonist selective for
m-opioid receptor.
bSNC80 ([(+)-4-[(aR)-a-(2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)-3-
methoxylbenzyl]-N,N0-diethyl benzamide), agonist selective for d-
opioid receptor.
cU69,593 (trans-3,4-dichloro-N-methyl[2-(1-pyrrolidinyl)cyclohexyl]
benzene acetamide), agonist selective for k opioid receptor.
dData from ref 13.
Results and Discussion
Two longer-chain compounds [the N-phenylpropyl (15)
and N-phenylbutyl (17) analogues) in the 1R,5S series
had moderate affinity (Table 1) for the m-opioid receptor
(Ki=42 and 30 nM, respectively). However, the com-
pound with the longest chain, the N-phenylpentyl ana-
logue in the 1R,5S or the 1S,5R series (19 and 20, Chart
1), had poor affinity for any opioid receptor (Ki>200
nM). The 10R and 10S methyl compounds in the 1R,5S
series (25 and 26), and the 20S hydroxyl compound in
the 1R,5S series (21), were also found to interact with
moderate affinity with the m-opioid receptor (Ki=29- 31
nM), comparable to the affinity of the known1 1S,5R-
(+)-5-m-hydroxyphenyl-N-methylmorphan (3, Ki=35
nM). None of the compounds had even moderate (Ki
<100 nM) affinity for d and k receptors, with the excep-
tion of 26 [(1R,5S,10S)-(+)-3-[2-(10-methyl-20-pheny-
lethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol, Ki=50 nM at
k], 23 [(1S,5R,20R)-(ꢀ)-3-[2-(20-hydroxy-20-phenylethyl)-
2-azabicyclo[3.3.1]non-5-yl]-phenol, Ki=70 nM at k],
and 27 [(1S,5R,10S)-(+)-3-[2-(10-methyl-20-phenylethyl)-
2-azabicyclo[3.3.1]non-5-yl]-phenol, Ki=90 nM at k].
Four compounds, 15, 17, 21, and 25 (Chart 1), had
some selectivity for the m-receptor (d/m =17, 23, 18, and
10, respectively), comparable to, or better than, the for-
merly reported13 1R,5S-(ꢀ)-2-phenethyl-5-(3-hydroxy-
phenyl)morphan (12) (d/m=13), and a relatively high
affinity compound, the 10S methyl compound in the
Replacement of an N-methyl substituent with an N-allyl
or N-cyclopropylmethyl is known to convert opioid
agonists, such as morphine and the 6,7-benzomorphan,
metazocine, to their respective antagonists, nalorphine
and cyclazocine. These N-substituents have different
effects in the m-hydroxyphenylmorphan series.6 Simi-
larly, when an N-substituent like N-phenylethyl replaces
an N-methyl moiety in the morphinans and 6,7-benzo-
morphans, compounds are obtained with potent agonist
activity. This also does not appear to be true for 5-phe-
nylmorphans. Some of our examined N-phenylethyl
compounds and their analogues in the 5-phenylmorphan
series were found to have opioid antagonist activity. We
have reported that (1R,5S)-(ꢀ)-5-(3-hydroxyphenyl)-N-
phenylethylmorphan (12) and its enantiomer 13, are
pure opioid antagonists.13 We now find that several
additional analogues, (1R,5S)-(ꢀ)-3-[2-(30-phenylpro-
pyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (15), (1R,5S)-
(ꢀ)-3-[2-(40 -phenylbutyl)-2-azabicyclo[3.3.1]non-5-yl]-
phenol (17), (1R,5S,20S)-(+)-3-[2-(20-hydroxy-20-phenyl-
ethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (21), (1R,5S,

A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
3325
10R)-(ꢀ)-3-[2-(10 -methyl-20 -phenylethyl)-2-azabicy-
Experimental
clo[3.3.1]non-5-yl]-phenol (25), and (1R,5S,10S)-(+)-3-
[2-(10-methyl-20-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-
phenol (26) are pure antagonists (as determined by the
[35S]GTPgS assay). Compound 15 was found to be the
most potent antagonist. It was about half as potent as
the potent opioid antagonist naloxone, at the m recep-
tor, 17 was about 0.2–0.25 times as potent as naloxone
at m, and 26 was ca. 0.15 times as potent as naloxone at
m. Compounds 21 and 25 are less potent antagonists.
The m-opioid affinity of these compounds was not in
accord with their efficacy in the [35S]GTPgS assay. As
seen in Table 2, the 1R,5S-N-phenylpropyl compound
15 had the highest efficacy. However, as seen in Table 1,
15 had the lowest affinity of all of the compounds listed
in Table 2. Efficacy and m-opioid affinity were more in
accord in the 1S,5R-N-phenylethyl compound 13 and
the 1R,5S-N-phenylbutyl compound 17. The N-pheny-
lethyl compound 12 had the highest m-opioid affinity,
but the affinity of its enantiomer 13 was similar to the
affinity of the 1R,5S-N-phenylbutyl compound 17, the
1R,5S-N-phenyl-20S-hydroxyethyl compound 21, and
the 1R,5S-N-phenyl-20R-methyl compound 25. The
marked structural differences in these compounds make
structure–activity analyses difficult. It is apparent that
compounds in the 1R,5S-series were better able to
interact with m-opioid receptors than their structural
relatives were in the 1S,5R-series (Table 1). However,
one 1S,5R compound, the-N-phenylethyl compound 13,
had good affinity (Ki=27 nM) for the m-opioid receptor,
as good or better than many of the compounds in the
1R,5S series. Thus, the distinguishing features that
increase efficacy or binding affinity at the m-opioid
receptor are not apparent.
Melting points were determined on a MEL-TEMP II
capillary melting-point apparatus and are uncor-
rected. Proton nuclear magnetic resonance (1H NMR)
spectra of the bases were recorded in CDCl3 with
tetramethylsilane (TMS) as the internal standard on a
Varian Gemini-300 spectrometer. Mass spectra (MS)
were recorded on a VG 7070E spectrometer or a
Finnigan 4600 spectrometer in the chemical ionization
mode (MS, CI-NH3). Thin-layer chromatography
(TLC) was performed on Analtech silica gel GHLF
0.25 mm plates. Flash column chromatography was
performed with Fluka silica gel 60 (mesh 220–240).
Elemental analyses were performed by Atlantic
Microlabs, Inc., Norcross, GA, USA and the results
were within 0.4% of the theoretical values unless
otherwise indicated.
General procedure for compounds 12, 13 and 15–24
(1R,5S)-(ꢀ)-3-[2-(20-Phenylethyl)-2-azabicyclo[3.3.1]non-
5-yl]-phenol (12). A mixture of (1R,5S)-(ꢀ)-3-[2-methyl-
2-azabicyclo [3.3.1]non-5-yl]-phenol (2) (1.5 g, 6.5
mmol) and acetic anhydride (20 mL) was heated at
90 ꢁC for 1 h. The mixture was evaporated. The residue
was diluted with chloroform (50 mL). The solution was
made basic with sodium carbonate (pH 9) and extracted
with chloroform. The combined organic solution was
dried over sodium sulfate, and evaporated. The residue
was taken up in toluene (30 mL), sodium bicarbonate
(168 mg, 2.0 mmol) and 2,2,2-trichloroethyl chlor-
oformate (1.78 mL, 12.9 mmol) were added, and the
mixture was refluxed for 15 h, and the solvent removed.
Chloroform (50 mL) was added to the residual material.
The solution was extracted with aqueous sodium car-
bonate (pH 9), and the aqueous layer was extracted with
chloroform. The combined organic solution was dried
over sodium sulfate, and the solvent removed. Zinc (2.1
g) was added to a solution of the residue in aqueous
acetic acid [glacial acetic acid (6 mL)+H2O (2 mL)] at
room temperature. After 3 h, concd ammonium
hydroxide solution (20 mL) was slowly added, and the
aqueous layer was extracted with chloroform. The
organic material was collected, dried over sodium sul-
fate and evaporated. Sodium bicarbonate (1.09 g, 13
mmol) and 1-bromo-2-phenylethane (0.93 mL, 6.8
mmol) were added to a DMF(50 mL) solution of the
residue and heated at 80 ꢁC for 2 h. The reaction mix-
ture was extracted with ethyl acetate, dried, and solvent
removed. The residue was dissolved in 10% NaOH/
methanol solution at 0 ꢁC. A saturated aqueous ammo-
nium chloride solution was added to the reaction mix-
ture after 10 min; it was extracted with ethyl acetate,
dried, and solvent removed. The residue was purified by
silica gel chromatography to yield 750 mg (36%) mp
The phenolic ring in the analogues which we prepared
was not sterically fixed, unlike the phenolic ring in the
6,7-benzomorphans, and morphinans, nor was it steri-
cally hindered, as it was in the 9b-methyl-5-phenylmor-
phans prepared by Thomas et al. Yet several of our N-
phenylethyl analogues were found to be opioid antago-
nists with moderate affinity and selectivity for the m-
opioid receptor. It is clearly not necessary to hinder the
rotation of the phenolic ring to obtain reasonably
potent opioid antagonists, confirming our initial
observation.13 However, it is likely that rotational
hindrance of the phenolic ring may improve receptor
affinity, and antagonist potency in the [35S]GTPgS
assay. The question of what the interaction is between
these ligands and their receptor that might promote
differentiation between opioid agonist and antagonist
activity is still unanswerable. The structure–activity
relationships that are apparent in our series of com-
pounds indicate that the spatial position of the phenyl
ring does not fully determine an opioid’s agonist or
antagonist actions. The major effect, a change from
agonist to antagonist, which can be obtained by a relatively
simple change in substitution on the amine function in
an opioid continues to be intriguing, and unexplained.
Perhaps these data, combined with data from future
structural modification, will provide the necessary
information to gain insight into the conversion of
opioid agonists to antagonists.
20
140–140.5 ꢁC; ½aꢂD ꢀ13.3ꢁ (c 0.6, MeOH). 1H NMR
(300 MHz, CDCl3) d 1.40 (1H, m), 1.66 (2H, m), 1.83–
2.15 (5H, m), 2.17 (2H, bd, J=11.7), 2.50–3.50 (1H, bs),
2.84 (4H, m), 3.00–3.08 (2H, m), 3.27 (1H, m), 6.65 (1H,
dd, J=2.1, 7.8), 6.81 (1H, d, J=1.8), 6.89 (1H, d,
J=7.8), 7.15–7.32 (6H, m). CIMS (NH3) m/z 322
[M+1]+. Anal. (C22H27NO) C, H, N.

3326
A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
0
-Phenylpentyl) -2 -azabicy-
clo[3.3.1]non-5-yl]-phenol (20). Free base (as a thick oil),
(1S,5R)-(+)-3-[2-(20-Phenylethyl)-2-azabicyclo[3.3.1]-
(1S,5R) -(+) -3 -[2 -(5
20
non-5-yl]-phenol (13). Free base, mp 140.0 ꢁC; ½aꢂD
20
ꢁ
+13.9 (c 0.5, MeOH). H NMR (300MHz, CDCl3) d
1.40 (1H, m), 1.66 (2H, m), 1.83–2.15 (5H, m), 2.17 (2H,
bd, J=11.7), 2.50–3.50 (1H, bs), 2.84 (4H, m), 3.00–3.08
(2H, m), 3.27 (1H, m), 6.65 (1H, dd, J=2.1, 7.8), 6.81 (1H,
d, J=1.8), 6.89 (1H, d, J=7.8), 7.15–7.32 (6H, m). CIMS
(NH3) m/z 322 [M+1]+. Anal. (C22H27NO) C, H, N.
½aꢂD +4.4ꢁ (c 0.1, MeOH). 1H NMR (300 MHz,
1
CDCl3) d 0.90 (1H, m), 1.00–3.00 (1H, bs), 1.35 (2H,
m), 1.62 (6H, m), 1.87 (3H, m),1.99 (2H, d, J=11.7),
2.13 (2H, d, J=12.6), 2.58 (4H, m), 2.94 (2H, m), 3.17
(1H, s), 6.60 (1H, dd, J=1.8, 7.8), 6.78 (1H, s), 6.87
(1H, d, J=7.8), 7.13–7.30 (6H, m). HRMS (FAB) m/z
364.635 (M+H)+, C25H34NO requires 364.2640. Anal.
.
(C25H33 NO 1/5H2O) C, H, N.
(1R,5S)-(ꢀ)-3-[2-(30-Phenylpropyl)-2-azabicyclo[3.3.1]-
20
non-5-yl]-phenol (15). Free base, mp 45–46 ꢁC; ½aꢂD
ꢁ
1
ꢀ7.8 (c 0.2, MeOH). H NMR (300 MHz, CDCl3) d
0.85 (1H, m), 1.26 (2H, m), 1.62 (2H, m), 1.83–2.16 (7H,
m), 2.65 (4H, m), 2.93 (2H, m), 3.18 (1H, s), 3.50–4.50
(1H, bs), 6.63 (1H, dd, J=2.1, 8.1), 6.78 (1H, d, J=1.8),
6.84 (1H, d, J=7.8), 7.1571–7.30 (6H, m). HRMS
(FAB) m/z 336.2321 (M+H)+, C23H30NO requires
(1R,5S,20S)-(+)-3-[2-(2 0 -Hydroxy-20 -phenylethyl)-2-
azabicyclo[3.3.1]non-5-yl]-phenol (21). Free base, mp 58–
20
59ꢁC; ½aꢂ +21.9ꢁ (c 0.09, MeOH). 1H NMR (300MHz,
CDCl3) dD1.47 (1H, m), 1.50–2.50 (1H, bs), 1.67 (2H, m),
1.88–2.19 (8H, m), 2.53 (1H, t, J=10.8), 2.84 (1H, dd,
J=3.0, 12.9), 3.11 (3H, m), 4.71 (1H, dd, J=3.0, 10.8),
6.67 (1H, dd, J=3.0, 7.8), 6.84 (1H, t, J=1.8), 6.93 (1H, d,
J=7.8), 7.20 (1H, t, J=7.8), 7.29–7.42 (5H, m). HRMS
(FAB) m/z 338.2112 (M+H)+, C22H28NO2 requires
.
336.2327. Anal. (C23H29NO 1/5H2O) C, H, N.
0
.
(1S,5R) -(+) -3 -[2 -(3
-Phenylpropyl) -2 -azabicy-
338.2120. Anal. (C22H27NO2 3/4H2O) C, H, N.
clo[3.3.1]non-5-yl]-phenol (16). Free base, mp 45–46 ꢁC;
20
½aꢂD +8.0ꢁ (c 0.2, MeOH). 1H NMR (300 MHz,
CDCl3) d 0.85 (1H, m), 1.26 (2H, m), 1.50–3.50 (1H,
bs), 1.61 (2H, m), 1.83–2.16 (7H, m), 2.65 (4H, m), 2.93
(2H, m), 3.17 (1H, s), 6.63 (1H, dd, J=1.8, 7.8), 6.80
(1H, d, J=1.8), 6.87 (1H, d, J=7.8), 7.12–7.30 (6H, m).
HRMS (FAB) m/z 336.2323 (M+H)+. C23H30NO
(1R,5S,20R)-(ꢀ)-3-[2-(20-Hydroxy-20-phenylethyl)-2-
azabicyclo[3.3.1]non-5-yl]-phenol (22). Free base, mp
20
154–155 ꢁC; ½aꢂD ꢀ31.9ꢁ (c 0.06, MeOH). 1H NMR
(300MHz, CDCl3) d 1.47–1.76 (3H, m), 1.88–2.14 (8H,
m), 2.44 (1H, t, J=10.5), 2.78 (1H, m), 3.03 (1H, dd,
J=3.9, 12.6), 3.24–3.33 (2H, m), 4.72 (1H, dd, J=3.0,
10.8), 6.67 (1H, dd, J=3.0, 7.8), 6.82 (1H, t, J=1.8), 6.91
(1H, d, J=7.8), 7.20 (1H, t, J=7.8), 7.25–7.41 (6H, m).
.
requires 336.2327. Anal. (C23H29NO 1/5H2O), C, H, N.
(1R,5S)-(ꢀ)-3-[2-(40-Phenylbutyl)-2-azabicyclo[3.3.1]-
non-5-yl]-phenol (17). Free base, mp 123.5–124.5 ꢁC;
CIMS (NH3) m/z 338 [M+1] . Anal. (C22H27NO2 1/
2H2O) C, H, N.
+
.
20
½aꢂD ꢀ6.7ꢁ (c 0.2, MeOH). 1H NMR (300 MHz, CDCl3)
d 0.90 (1H, m), 1.29 (2H, m), 1.50–3.00 (1H, bs), 1.64
(4H, m), 1.83–2.14 (7H, m), 2.63 (4H, m), 2.93 (2H, m),
3.15 (1H, s), 6.63 (1H, dd, J=1.8, 7.8), 6.80 (1H, s), 6.88
(1H, d, J=9.0), 7.13–7.30 (6H, m). HRMS (FAB) m/z
350.2477 (M+H)+. C24H32NO requires 350.2484.
Anal. (C24H31NO) C, H, N.
(1S,5R,20R)-(ꢀ)-3-[2-(2 0 -Hydroxy-20 -phenylethyl)-2-
azabicyclo[3.3.1]non-5-yl]-phenol (23). Free base, mp 58–
20
59 ꢁC; ½aꢂD ꢀ23.0ꢁ (c 0.06, MeOH). 1H NMR
(300 MHz, CDCl3) d 1.47 (1H, m), 1.50–2.50 (1H, bs),
1.67 (2H, m), 1.88–2.19 (8H, m), 2.54 (1H, t, J=10.8),
2.85 (1H, dd, J=3.9, 12.6), 3.11 (3H, m), 4.72 (1H, dd,
J=3.0, 10.8), 6.67 (1H, dd, J=3.0, 7.8), 6.84 (1H, t,
J=1.8), 6.93 (1H, d, J=7.8), 7.20 (1H, t, J=7.8), 7.29–
7.42 (5H, m). HRMS (FAB) m/z 338.2119 (M+H)+,
(1S,5R)-(+)-3-[2-(40-Phenylbutyl)-2-azabicyclo[3.3.1]-
non-5-yl]-phenol (18). Free base, mp 123.5–124.5 ꢁC;
20
½aꢂD +6.0ꢁ (c 0.1, MeOH). 1H NMR (300 MHz,
C22H28NO2 requires 338.2120. Anal. (C22H27NO2 3/
4H2O) C, H, N.
.
CDCl3) d 0.90 (1H, m), 1.29 (2H, m), 1.50–3.00 (1H,
bs), 1.63 (4H, m), 1.83–2.14 (7H, m), 2.63 (4H, m), 2.93
(2H, m), 3.15 (1H, t, J=3.0), 6.63 (1H, dd, J=1.8, 7.8),
6.80 (1H, s), 6.88 (1H, d, J=7.8), 7.13–7.30 (6H, m).
HRMS (FAB) m/z 350.2476 (M+H)+. C24H32NO
(1S,5R,20S)-(+)-3-[2-(2 0 -Hydroxy-20 -phenylethyl)-2-
azabicyclo[3.3.1]non-5-yl]-phenol (24). Free base, mp
20
.
requires 350.2484. Anal. (C24H31NO 1/5H2O) C, H, N.
154–155 ꢁC; ½aꢂD +31.6ꢁ (c 0.07, MeOH). 1H NMR
(300 MHz, CDCl3) d 1.43–1.75 (3H, m), 1.88–2.16 (8H,
m), 2.44 (1H, t, J=10.8), 2.77 (1H, m), 3.03 (1H, dd,
J=3.9, 13.8), 3.23–3.33 (2H, m), 4.72 (1H, dd, J=3.0,
10.8), 6.66 (1H, dd, J=2.1, 7.8), 6.82 (1H, t, J=2.1), 6.90
(1H, d, J=7.8), 7.19 (1H, t, J=7.8), 7.25–7.41 (5H, m).
HRMS (FAB) m/z 338.2114 (M+H)+, C22H28NO2
(1R,5S)-(ꢀ)-3-[2-(50-Phenylpentyl)-2-azabicyclo[3.3.1]-
20
non-5-yl]-phenol (19). Free base (as a thick oil), ½aꢂD ꢀ5.2ꢁ
1
(c 0.2, MeOH). H NMR (300MHz, CDCl3) d 0.90 (1H,
m), 1.35 (2H, m), 1.62 (6H, m), 1.87 (3H, m), 1.99 (2H, d,
J=13.5), 2.13 (2H, d, J=12.6), 2.58 (4H, m), 2.93 (2H, m),
3.19 (1H, s), 3.90–5.20 (1H, bs), 6.60 (1H, dd, J= 2.1, 7.8),
6.75 (1H, s), 6.82 (1H, d, J=7.8), 7.13–7.30 (6H, m).
HRMS (FAB) m/z 364.2633 (M+H) +, C25H34NO
.
requires 338.2120. Anal. (C22H27NO2 1/10H2O) C, H, N.
(1R,5S,10R)-(ꢀ)-3-[2-(10-Methyl-20-phenylethyl)-2-aza-
.
requires 364.2640. Anal. (C25H33NO 1/3H2O) C, H, N.
bicyclo[3.3.1]non-5-yl]-phenol (25).
A
mixture of

A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
3327
(1R,5S)-(ꢀ)-3-[2-methyl-2-azabicyclo[3.3.1]non-5-yl]-
phenol (2) (0.5 g, 2.2 mmol) and acetic anhydride (5 mL)
was heated at 90ꢁC for 1 h. Solvent was removed, and
chloroform (20 mL) was added. The solution was washed
with aqueous sodium carbonate (pH 9). The aqueous
layer was extracted with chloroform, and the combined
organic solutions were dried over sodium sulfate and
solvent removed. The residue was dissolved in toluene
(20 mL), sodium bicarbonate (56 mg, 0.7 mmol) and
2,2,2-trichloroethyl chloroformate (0.6 mL, 4.4 mmol)
were added, and the mixture was refluxed for 15 h. After
removal of solvent, the residue was dissolved in chloro-
form (30 mL). The solution was washed with aqueous
sodium carbonate (pH 9). The aqueous layer was extrac-
ted with chloroform. The combined organic solutions
were dried over sodium sulfate, and solvent removed.
The residue was purified by silica gel chromatography.
To a solution of the purified compound in aqueous
acetic acid was added 0.7 g of zinc at room temperature.
After 3 h, a sodium bicarbonate solution was slowly
added, and the aqueous layer was extracted with
chloroform. The combined organic material was dried
over sodium sulfate and solvent removed. The residue
was dissolved in 20 mL of EtOH and 2 mL of ammo-
nium hydroxide solution was added. The reaction mix-
ture was stirred overnight at room temperature, and
solvent removed. The residue was dissolved in benzene
(30 mL), 1-phenyl-propan-2-one (350 mg, 1.2 equiv)
and TsOH (0.01 g) were added, and water removed
from the mixture by refluxing overnight with a Dean–
Stark trap. Solvent was removed from the reaction
mixture. The residue was dissolved in MeOH and
NaBH3CN (136 mg, 1.0 mol equiv) and two drops of
acetic acid were added. The reaction mixture was stirred
for 3 h at room temperature and solvent was removed.
The residue was dissolved in chloroform and washed
with aqueous sodium bicarbonate, dried, and solvent
removed. The residue was purified by silica gel chroma-
tography to afford a diastereomeric mixture (238 mg,
33%). The free base was dissolved in 10 mL of EtOH
and 1 mL of 48% HBr was added. Solvent was removed
from the mixture. The residue was dissolved in 2 mL of
EtOH and after standing overnight at 0 ꢁC, gave 50 mg
of the crystalline HBr salt of 25 (5.6%) mp 287–288 ꢁC;
(3H, d, J=6.3), 1.30–1.43 (2H, m), 1.46–1.74 (3H, m),
1.85–2.00 (3H, m), 2.00–2.20 (3H, m), 2.45 (1H, dd,
J=9.3, 12.9), 2.88–3.10 (4H, m), 3.29 (1H, m), 6.65 (1H,
dd, J=1.8, 7.8), 6.80 (1H, t, J=2.1), 6.89 (1H, d, J=7.5),
7.14–7.32 (6H, m). CIMS (NH3) m/z 334 [Mꢀ1]+. Anal.
.
(C23H29NO HBr) C, H, N; X-ray structure analysis
unequivocally determined the structure.
(1S,5R,10S)-(+)-3-[2-(10-Methyl-20-phenylethyl)-2-aza-
bicyclo[3.3.1]non-5-yl]-phenol (27). This material was
prepared using the procedure described for 25: mp 287–
20
288 ꢁC HBr salt; ½aꢂD +15.1ꢁ (c 0.05, MeOH, HBr salt).
1H NMR (300 MHz, CDCl3) d 1.04 (3H, d, J=6.3),
1.12–1.28 (2H, m), 1.36–1.52 (2H, m), 1.63 (2H, d,
J=11.7), 1.80–2.24 (5H, m), 2.49 (1H, dd, J=8.4, 12.9),
2.82–2.93 (2H, m), 2.95–3.10 (2H, m), 3.24 (1H, m), 6.64
(1H, dd, J=2.4, 8.1), 6.78 (1H, t, J=2.4), 6.87 (1H, d,
J=8.1), 7.14–7.32 (6H, m). CIMS (NH3) m/z 3346
+
.
[M+1] . Anal. (C23H29NO HBr) C, H, N.
(1S,5R,10R)-(ꢀ)-3-[2-(10-Methyl-20-phenylethyl)-2-aza-
bicyclo[3.3.1]non-5-yl]-phenol (28). This material was
prepared using the procedure described for 26: mp 236–
20
237 ꢁC HBr salt; ½aꢂD ꢀ11.2ꢁ (c 0.09, MeOH, HBr salt). 1H
NMR (300 MHz, CDCl3) d 1.05 (3H, d, J=6.3), 1.30–1.43
(2H, m), 1.46–1.74 (3H, m), 1.85–2.00 (3H, m), 2.00–2.20
(3H, m), 2.45 (1H, dd, J=8.7, 12.6), 2.88–3.10 (4H, m),
3.28 (1H, m), 6.65 (1H, dd, J=2.4, 8.1), 6.81 (1H, t,
J=1.8), 6.90 (1H, d, J=7.5), 7.14–7.32 (6H, m). CIMS
+
.
(NH3) m/z 336 [M+1] . Anal. (C23H29NO HBr) C, H, N.
0
(1R,5S)-(ꢀ)-4-Bromo-3-[2-(2-phenylethyl)-2-azabicyclo
[3.3.1]non-5-yl]-phenol (29). A mixture of (1R,5S)-(ꢀ)-3-
[2-(2-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol
(12) (50 mg, 0.16 mmol) in 10 mL of CS2 was cooled to
0 ꢁC and Br2 (8.09 mL, 1.01 equiv) in CS2 was added.
The solution was evaporated and the residue was dis-
solved in chloroform and washed with aqueous sodium
bicarbonate. The combined organic solutions were dried
over sodium sulfate and evaporated. The residue was
purified by silica gel chromatography to afford 53 mg of
29 (85%) as the free base. The free base was dissolved in
chloroform, 48% HBr aq added, and solvent removed.
The residue was recrystallized in MeOH to afford 22 mg
20
½aꢂD ꢀ14.8ꢁ (c 0.07, MeOH, HBr salt). 1H NMR
(300 MHz, CDCl3) d 1.04 (3H, d, J=6.3), 1.12–1.28
(2H, m), 1.36–1.52 (2H, m), 1.62 (2H, d, J=12.3), 1.80–
2.24 (5H, m), 2.49 (1H, dd, J=8.1, 12.9), 2.82–2.93
(2H, m), 2.95–3.10 (2H, m), 3.23 (1H, m), 6.64 (1H,
dd, J=2.1, 8.1), 6.79 (1H, t, J=2.1), 6.88 (1H, d,
J=7.8), 7.14–7.32 (6H, m). CIMS (NH3) m/z 334
20
(29%) mp 282–283 ꢁC (dec) HBr salt; ½aꢂD ꢀ12.7ꢁ (c
1
0.07, MeOH, HBr salt). H NMR (300 MHz, CDCl3) d
1.39 (1H, m), 1.50–1.76 (2H, m), 1.82–2.22 (8H, m), 2.83
(4H, m), 3.03 (2H, m), 3.24 (1H, m), 6.81 (1H, dd,
J=1.8, 8.7), 7.02 (1H, d, J=2.1), 7.18–7.34 (5H, m),
7.38 (1H, d, J=8.7). CIMS (NH3) m/z 400 [M+1]+.
+
.
[Mꢀ1] . Anal. (C23H29NO HBr) C, H, N; X-ray
structure analysis unequivocally determined the mole-
cular structure.
.
Anal. (C22H26BrNO HBr) C, H, N.
0
(1R,5S,10S)-(+)-3-[2-(10-Methyl-20-phenylethyl)-2-aza-
bicyclo[3.3.1]non-5-yl]-phenol (26). The residual solution
of 25 was evaporated and to the residue was added 1
mL of MeOH and the solution was allowed to stand
overnight at 0 ꢁC. Crystalline HBr salt 26 was obtained
(1S,5R)-(+)-4-Bromo-3-[2-(2-phenylethyl)-2-azabicyclo
[3.3.1]non-5-yl]-phenol (30). This material was prepared
using the procedure described for 29.: mp 282–283 ꢁC
20
(dec.) HBr salt; ½aꢂD +13.4ꢁ (c 0.06, MeOH, HBr salt). 1H
NMR (300MHz, CDCl3) d 1.39 (1H, m), 1.50–1.76 (2H,
m), 1.82–2.22 (8H, m), 2.83 (4H, m), 3.03 (2H, m), 3.25
(1H, m), 6.81 (1H, dd, J=1.8, 8.7), 7.02 (1H, d, J=2.1),
20
(20 mg, 2.2%) mp 237–238 ꢁC; ½aꢂD +12.0ꢁ (c 0.08,
1
MeOH, HBr salt). H NMR (300 MHz, CDCl3) d 1.05

3328
A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
7.18–7.34 (5H, m), 7.38 (1H, d, J=8.7). CIMS (NH3) m/z
+
phenol (2) (0.5 g, 2.2 mmol) and acetic anhydride (5
mL) was heated at 90 ꢁC for 1 h. Solvent was removed,
and chloroform (20 mL) was added. The solution was
washed with aqueous sodium carbonate (pH 9). The
aqueous layer was extracted with chloroform, and the
combined organic solutions were dried over sodium
sulfate and solvent removed. The residue was dissolved
in toluene (20 mL), sodium bicarbonate (56 mg, 0.7
mmol) and 2,2,2-trichloroethyl chloroformate (0.6 mL,
4.4 mmol) were added, and the mixture was refluxed for
15 h. After removal of solvent, the residue was dissolved
in chloroform (30 mL). The solution was washed with
aqueous sodium carbonate (pH 9). The aqueous layer
was extracted with chloroform. The combined organic
solutions were dried over sodium sulfate, and solvent
removed. The residue was purified by silica gel chroma-
tography. To a solution of the purified compound in
aqueous acetic acid was added 0.7 g of zinc at room
temperature. After 3 h, a sodium bicarbonate solution
was slowly added, and the aqueous layer was extracted
with chloroform. The combined organic material was
dried over sodium sulfate and solvent removed. The
residue was dissolved in 20 mL of EtOH and 2 mL of
ammonium hydroxide solution was added. The reaction
mixture was stirred overnight at room temperature, and
solvent removed. The residue was added (R)-(ꢀ)-2-phe-
.
400 [M+1] . Anal. (C22H26BrNO HBr) C, H, N.
(1R,5S)-(ꢀ)-2-Bromo-5-[2-(20-phenylethyl)-2-azabicyclo
[3.3.1]non-5-yl]-phenol (31). To a solution of (1R,5S)-(ꢀ)
-3-[2-(20-phenylethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol
(12) (100 mg, 0.31 mmol) in 10 mL of DMSO was added
0.21 mL (6.0 equiv) of 48% HBr aq at room temperature
The reaction mixture was stirred for 24 h, neutralized with
sodium bicarbonate solution, and extracted with ethyl
acetate. The organic solution was dried over sodium sul-
fate and evaporated. The residue was purified by silica gel
chromatography. The free base was dissolved in chloro-
form, 48% HBr aq added, and solvent removed to give
120.5 mg, of the HBr salt of 31 (80%) The residue was
recrystallized in methanol to afford 58 mg (47%) mp 280–
20
281 ꢁC (dec.) HBr salt; ½aꢂD ꢀ13.8ꢁ (c 0.07, MeOH, HBr
salt). 1H NMR (300MHz, CDCl3) d 1.39 (1H, m), 1.50–
1.76 (2H, m), 1.82–2.08 (5H, m), 2.15 (1H, m), 2.50–4.00
(1H, bs), 2.85 (4H, m), 3.03 (2H, m), 3.24 (1H, m), 6.76
(1H, dd, J=3.0, 7.8), 7.00 (1H, d, J=2.1), 7.18–7.34 (5H,
m), 7.37 (1H, d, J=8.7). CIMS (NH3) m/z 400
+
.
[M+1] . Anal. (C22H26BrNO HBr) C, H, N.
0
(1S,5R)-(+)-2-Bromo-5-[2-(2-phenylethyl)-2-azabicyclo
nylpropionic acid (0.32 mL), DMF(10 mL), Et N (0.36
3
[3.3.1]non-5-yl]-phenol (32). This material was prepared
using the procedure described for 31: mp 280–281 ꢁC
mL), DPPA (0.56 mL) and then stirred for 10 h at room
temperature. The reaction mixture was added water
(100 mL) and extracted AcOEt (20 mL X3). The
organic layer was dried over Na2SO4 and then evapo-
rated. The residue was purified by silica gel chromato-
graphy. The purified amide in THF(20 mL) was added
BH3–THFcomplex (1 M/THF, 4.3 mL) and refluxed
for 3 days. The reaction mixture was evaporated under
reduced pressure. The residue was added MeOH/HCl
aq (1/1, 20 mL) and then stirred 2 h at room tempera-
ture The reaction mixture was evaporated and then
residue was extracted NaHCO3 aq/AcOEt and dried,
evaporated. The residue was purified by silica gel chro-
matography. The free base was dissolved in 10 mL of
EtOH and 1 mL of 48% HBr was added. Solvent was
removed from the mixture. The residue was dissolved in
1 mL of MeOH and after standing overnight at 0 ꢁC,
gave 162 mg of the crystalline HBr salt of 34 (18%) mp
20
(dec.) HBr salt; ½aꢂD +12.5ꢁ (c 0.05, MeOH, HBr salt).
1H NMR (300 MHz, CDCl3) d 1.41 (1H, m), 1.50–1.76
(2H, m), 1.82–2.08 (5H, m), 2.15 (1H, d, J=12.6), 2.85
(4H, m), 3.03 (2H, dd, J=3.0, 5.7), 3.30 (1H, m), 4.00–
5.50 (1H, bs), 6.74 (1H, dd, J=3.0, 8.7), 6.97 (1H, d,
J=1.8), 7.18–7.34 (5H, m), 7.37 (1H, d, J=8.7). CIMS
(NH3) m/z 400 [M+1]+ Anal. (C22H26BrNO HBr) C,
.
H, N; X-ray structure analysis unequivocally deter-
mined the molecular structure.
0
(Æ)-2,4-Dibromo-5-[2-(2-phenylethyl)-2-azabicyclo[3.3.1]
non-5-yl]-phenol (33 ). A mixture of (Æ)-3-[2-(2-pheny-
lethyl)-2-azabicyclo[3.3.1]non-5-yl]-phenol (300 mg, 1.3
mmol) in 30 mL of CS2 was cooled to 0 ꢁC and Br2 (0.2
mL, 3.0 equiv) was added. The solution was evaporated
and the residue was extracted with chloroform and
aqueous sodium bicarbonate. The organic layer was
dried over sodium sulfate, solvent removed, and the
residue was purified by silica gel chromatography. The
free base was dissolved in chloroform, 48% aq HBr
added, and solvents removed. The residue was recrys-
20
ꢁ
123–124 ꢁC; ½aꢂD +16.2 (c 0.2, MeOH, HBr salt). H
1
NMR (300 MHz, CDCl3) d 1.29 (3H, d, J=6.9), 1.62
(2H, m), 1.77–2.09 (7H, m), 2.53–2.63 (1H, m), 2.79–
3.07 (7H, m), 6.64 (1H, dd, J=2.4, 8.1), 6.79 (1H, t,
J=1.8), 6.87 (1H, d, J=8.1), 7.13–7.32 (6H, m). CIMS
(NH3) m/z 336 [M+1]+. Anal. (C23H29NO 1/
.
.
.
2H2O HBr) C, H, N.
tallized in MeOH to afford 120 mg (17%) of 33 HBr:
mp 239–240 ꢁC (dec). H NMR (300 MHz, CDCl3) d
1
0.87 (1H, m), 1.10–2.10 (8H, m), 2.25 (1H, d, J=12.0),
2.80–3.10 (6H, m), 3.49 (1H, s), 2.40–3.80 (1H, bs), 6.86
(1H, s), 7.00 (1H, d, J=2.1), 7.22–7.36 (5H, m), 7.61
(1H, s). CIMS (NH3) m/z 480 [M+1]+. Anal. (C22H25
(1R,5S,20S)-(ꢀ)-3-[2-(20-Methyl-20-phenylethyl)-2-aza-
bicyclo[3.3.1]non-5-yl]-phenol (35). This material was
prepared using the procedure described for 34: mp 127–
20
128 ꢁC HBr salt; ½aꢂD ꢀ36.3ꢁ (c 0.2, MeOH, HBr salt).
.
Br2NO HBr) C, H, N.
1H NMR (300 MHz, CDCl3) d 1.27 (3H, d, J=6.9), 1.60
(2H, m), 1.77–2.09 (7H, m), 2.58–2.66 (1H, m), 2.72–3.08
(7H, m), 6.63 (1H, dd, J=2.1, 8.1), 6.79 (1H, t-like), 6.88
(1H, d, J=7.8), 7.13–7.32 (6H, m). CIMS (NH3) m/z 336
0
(1R,5S,20R)-(+)-3-[2-(2 -Methyl-2 0 -phenylethyl)-2-
azabicyclo[3.3.1]non-5-yl]-phenol (34). A mixture of
(1R,5S)-(ꢀ)-3-[2-methyl-2-azabicyclo[3.3.1]non-5-yl]-
+
.
[M+1] . Anal. (C23H29BrNO HBr) C, H, N.

A. Hashimoto et al. / Bioorg. Med. Chem. 10 (2002) 3319–3329
3329
Single crystal X-ray analyses
Robert Horel (NIDA, NIH) for his expert technical
assistance.
X-ray data for compounds 25, 26, and 32 were collected
on a computer controlled Bruker P4 diffractometer
using CuKa radiation and a graphite monochromator in
the incident beam. Data were corrected for Lorentz,
polarization and absorption effects. The structures were
solved by direct methods with the aid of program
SHELXTL17 and refined by full-matrix least-squares on
F2 values using program SHELXLS.17 The parameters
refined included the coordinates and anisotropic ther-
mal parameters for all non-hydrogen atoms. Hydrogen
atoms bonded to carbon atoms were included using a
riding model in which the coordinate shifts of the car-
bon atoms were applied to their attached hydrogens
References and Notes
1. May, E. L.; Murphy, J. G. J. Org. Chem. 1955, 20, 257.
2. May, E. L. Science 1973, 181, 407.
3. May, E. L. J. Med. Chem. 1980, 23, 225.
4. Eddy, N. B.; May, E. L. Science 1973, 181, 407.
5. May, E. L.; Murphy, J. G. J. Org. Chem. 1955, 20, 1197.
6. Ong, H.; Oh-ishi, T.; May, E. L. J. Med. Chem. 1974, 17, 133.
7. May, E. L.; Takeda, M. J. Med. Chem. 1970, 13, 805.
8. Thomas, J. B.; Zheng, X. L.; Mascarella, S. W.; Rothman,
R. B.; Dersch, C. M.; Partilla, J. S.; Flippen-Anderson, J. L.;
George, C. F.; Cantrell, B. E.; Zimmerman, D. M.; Carroll,
F . I.J. Med. Chem. 1998, 41, 4143.
9. Zimmerman, D. M.; Nickander, R.; Horng, J. S.; Wong,
D. T. Nature 1978, 275, 332.
10. Zimmerman, D. M.; Leander, J. D.; Cantrell, B. E.; Reel,
J. K.; Snoddy, J.; Mendelsohn, L. G.; Johnson, B. G.; Mitch,
C. H. J. Med. Chem. 1993, 36, 2833.
˚
with C–H=0.96 A. H angles were idealized and Uiso (H)
values were set at fixed ratios of Uiso values of their
bonded atoms. Coordinates were refined for hydrogens
involved in hydrogen bonds. 32 (C22H27NOBr2) crystal-
lized in the monoclinic space group P21 with a=13.581(1),
11. May, E. L. J. Org. Chem. 1956, 21, 899.
ꢁ
˚
b=10.046(1), c 15.384(2) A and b=90.44(1) and two
molecules per asymmetric unit. Final R-factors were
0.027 for 2942 observed data and 0.028 for all 3014
unique data. 25 (C23H30NOBr) crystallized in the triclinic
space group P1 with a=7.481(1), b=8.669(1), c=8.871(1)
12. Awaya, H.; May, E. L.; Aceto, M. D.; Merz, H.; Rogers,
M. E.; Harris, L. S. J. Med. Chem. 1984, 27, 536.
13. Hashimoto, A.; Coop, A.; Rothman, R. B.; Dersch, C.
M.; Xu, H.; Horel, R.; George, C.; Jacobson, A. E.; Rice, K.
C. In Problems of Drug Dependence 1999. Proceedings of the
61st. Annual Scientific Meeting; Harris, L. S., Ed.; Supt. of
Docs., U.S. Govt. Print. Off.: Washington, DC, 2000; NIDA
Research Monograph 180, p 250.
14. Bonjoch, J.; Casamitjana, J.; Bosch, J. Tetrahedron 1988,
44, 1735.
15. Bertha, C. M.; Mattson, M. V.; Flippen-Anderson, J. L.;
Rothman, R. B.; Xu, H.; Becketts, K.; Rice, K. C. J. Med.
Chem. 1994, 37, 3163.
ꢁ
˚
A, a=112.61(1), b=93.95(1) and g=98.83(1) . Final R-
factors were 0.026 for 1753 observed data and 0.027 for
all 1759 unique data. Compound 26 (C23H30NOBr)
crystallized in the orthorhombic space group P212121
˚
with a=9.716(1), b=9.897(1), and c=21.814(2) A.
Final R-factors were 0.034 for 2012 observed data and
0.035 for all 2074 unique data. Coordinates for all three
compounds have been deposited with the Cambridge
Crystallographic Data Centre (Cambridge University
Chemical Laboratory, Cambridge CB2 1EW, UK).
16. Majetich, G.; Hicks, R.; Reister, S. J. Org. Chem. 1997,
62, 43212.
17. Sheldrick, G. M. Version 5.03; Bruker Analytical X-ray
Instruments: Madison, WI, 1997.
18. Rothman, R. B.; Xu, H.; Seggel, M.; Jacobson, A. E.; Rice,
K. C.; Brine, G. A.; Carroll, F. I. Life Sci. 48, PL111. 1991.
19. Rothman, R. B.; de Costa, B. R.; Bykov, V.; Jacobson,
A. E.; Rice, K. C.; Brady, L. S. Peptides 1990, 11, 311.
20. Xu, H.; Hashimoto, A.; Rice, K. C.; Jacobson, A. E.;
Thomas, J. B.; Carroll, F. I.; Lai, J.; Rothman, R. B. Synapse
2001, 39, 64.
Biological assays. Details provided in refs 18–20.
Acknowledgements
We would like to thank the Office of Naval Research
and NIDA for financial support. We also thank Noel
Whittaker (NIDDK, NIH) for mass spectral data, and
21. Partilla, J. S.; Carroll, F. I.; Thomas, J. B.; Rice, K. C.;
Zimmerman, D. M.; Rothman, R. B. Analgesia 1999, 4, 27.