Probes for narcotic receptor mediated phenomena: Synthesis of rac-(3R,6aS,11aR)-1,3,4,5,6,11a-hexahydro-2-methyl- 2H-3,6a-methanobenzofuro[2,3-c]azocin-8-ol, an epoxy isomer of 5-phenylmorphan

Article abstract of DOI:10.1002/hlca.200390048

The synthesis of a series of epoxy 5-phenylmorphans is being explored in order to determine the conformational requirements of the phenolic ring in a phenylmorphan molecule that may be needed both for binding to a specific opioid receptor and for exhibiti

Full text of DOI:10.1002/hlca.200390048

484
Helvetica Chimica Acta Vol. 86 (2003)
Probes for Narcotic Receptor Mediated Phenomena
Part 301)
Synthesis of rac-(3R,6aS,11aR)-1,3,4,5,6,11a-Hexahydro-2-methyl-2H-3,6a-
methanobenzofuro[2,3-c]azocin-8-ol, an Epoxy Isomer of 5-Phenylmorphan
by Joannes T. M. Lindersa)2), Seid Mirsadeghia), Judith L. Flippen-Andersonb), Clifford Georgeb),
Arthur E. Jacobsona), and Kenner C. Rice*a)
a) Laboratory of Medicinal Chemistry, Building 8, Room B1-23, National Institute of Diabetes and Digestive
and Kidney Diseases, National Institutes of Health, Bethesda, Maryland, 20892-0815, USA
(phone: ‡1(301)983-1310; fax ‡1(301)4020589; e-mail: kr21f@nih.gov)
b) Laboratory for the Structure of Matter, Naval Research Laboratory, Washington, D.C. 20375, USA
The synthesis of a series of epoxy 5-phenylmorphans is being explored in order to determine the
conformational requirements of the phenolic ring in a phenylmorphan molecule that may be needed both for
binding to a specific opioid receptor and for exhibiting opioid agonist or antagonist activity. Of the twelve
possible ortho-and para-bridged isomers (a f) (Fig. 1), we now report the synthesis of the para-d isomer, rac-
(3R,6aS,11aR)-2-methyl-1,3,4,5,6,11a-hexahydro-2H-3,6a-methanobenzofuro[2,3-c]azocin-8-ol (3). Compound
3 was synthesized via construction of the 5-phenylazabicyclo[3.3.1]non-3-ene skeleton (Scheme 1) and
subsequent closure of the epoxy bridge (Scheme 2). As determined by an X-ray diffraction study, the epoxy
bridge, restricting the phenyl-ring rotation, fixed the dihedral angle between the least-squares planes through
the phenyl ring and atoms N(2), C(3), C(11a), and C(6a) of the piperidine ring (Fig. 2) at 43.08, and the torsion
angle C(12)ÀC(6a)ÀC(6b)ÀC(10a) at À95.08.
Introduction. The initial exploration of 5-phenylmorphans as analgesics was
initiated as part of the attempt at simplification of the multi-ring structure of the known
opioids at that time (e.g., the 4,5-epoxymorphinans) [2]. In the phenylmorphan series
of compounds, the aromatic ring is constrained to the equatorial position relative to the
piperidine ring, in contrast to the morphinans where the aromatic ring is in an axial
conformation. It was hoped that the removal of appendages from the more-complex
molecules might result in compounds with fewer side-effects than those usually
attributed to the morphine-type of opioids, such as tolerance, physical dependence
liability, and abuse potential, among many others.
Over the past five decades, May and co-workers have prepared a number of N-
substituted phenylmorphans, and some of them have been resolved [3][4]. An
interesting effect was noted with the enantiomers of 2-methyl-5-(3-hydroxyphenyl)-
morphan (ˆ 3-(2-methyl-2-azabicyclo[3.3.1]non-5-yl)phenol)) [5]. The racemic com-
pound 1a was a strong morphine-like agonist. The agonist effect is retained by both
1
)
)
Paper No. 29: [1].
2
Current address: Department of Medicinal Chemistry, Johnson & Johnson Pharmaceutical Research and
Development, Turnhoutseweg 30, B-2340 Beerse.

Helvetica Chimica Acta Vol. 86 (2003)
485
enantiomers, in contrast to the 4,5-epoxymorphinans, such as morphine, where only the
natural (À)-morphine absolute configuration has opioid receptor affinity or efficacy.
The (‡)-phenylmorphan enantiomer 1b, which is twice as potent as morphine in the
mouse hot plate assay, completely substituted for morphine in morphine-dependent
monkeys. However, its optical antipode 1c did not substitute for morphine [6]. The
(À)-phenylmorphan enantiomer, with about half of the antinociceptive potency of
morphine, had modest opioid antagonist properties. The phenolic OH group meta-
oriented to the piperidine ring was also noted to be essential for high antinociceptive
activity. The 9a-methyl phenylmorphan isomer 1d showed diminished agonist potency;
it displayed a mixed agonist antagonist profile [4]. Its antagonist activity was found to
be half that of nalorphine and was shown to reside only in the (‡)-enantiomer (which
has the same absolute configuration (1R,5S,9R) as 1c) [7].
We have, for some time [8 11], been involved in restricting the rotation of the
aromatic ring in phenylmorphans by using an epoxy-bridge, in order to study the
relationship between the conformation of the m-hydroxyphenyl ring and the
pharmacological profile of the compound. In theory, our approach allows a more
exact determination of the dihedral angle of the aromatic ring to the piperidine ring
needed for either opioid agonist or antagonist activity than an unbridged analogue.
Fig. 1 depicts the various epoxy-bridged compounds that can possibly be synthesized,
which would have a fixed, determinable angular relationship. There are twelve such
Fig. 1. Possible epoxy-bridged 5-phenylmorphans
(ortho or para (a f) isomers)

486
Helvetica Chimica Acta Vol. 86 (2003)
racemates that must be prepared, and, thus, 24 enantiomeric compounds with a
phenolic OH group meta to the piperidine ring.
Our theoretical idea for the determination of the proper angle needed for high
receptor affinity and efficacy is, then, dependent on our ability to synthesize these
compounds. Further, this determination would become practicable only if we were able
to obtain compounds with in vivo or in vitro opioid agonist or antagonist activity.
Although there can be no a priori assurance that an epoxy-bridged phenylmorphan
would have any opioid-like activity (lesser modifications of opioid structures with
potent in vivo activity have been known to eliminate such activity), it was reassuring to
note that Hutchinson et al. found that their epoxy-bridged compounds 2, with structural
similarity to our para-d phenylmorphan isomer 3, were pharmacologically active [12].
Restriction of the rotation of the aromatic ring in their phenylmorphan-like structures
led to compounds that were noted to have high affinity (2, R ˆ PhCH2CH2) and were
found to be reasonably potent antinociceptive compounds (opioid agonists) [12]. These
authors stated that some of their N-substituted compounds might have opioid
antagonist activity, but that was not examined nor were the racemic compounds
resolved.
Recently, we [13] and others [14] re-examined the 5-phenylmorphan structure from
a different point of view. Thomas et al. noted that pure opioid antagonists had not been
obtained in the phenylmorphan series and decided that this was due to unrestricted
rotation of the aromatic ring [15]. Restricting its rotation, with a 9b-Me substituent (i.e.
1e), gave ligands that were pure, nonselective, antagonists in the GTPgS assay,
especially when an N-phenylethyl substituent was present. This approach was based on
previous studies by Zimmerman et al. in the 4-phenylpiperidine series of analgesics
[16]. We have found that restriction of the rotation of the aromatic ring could not be the
primary mechanism in determining the opioid antagonist behavior of the compound in
the phenylmorphan series [13]. However, conformationally restricted phenylmorphan
opioid antagonists were found to have considerably higher receptor affinity [17] than
those with freely rotating phenolic rings [13].
Thus far, we have reported the synthesis of rac-(4R,6aR,11bR)-2,3,4,5,6,6a-
hexahydro-3-methyl-1H-4,11b-methanobenzofuro[3,2-d]azocine-8-ol (4) [8][9], the
ortho-a epoxy-bridged phenylmorphan isomer, rac-(1R,4aR,9aR)-1,3,4,9a-tetrahydro-
2-methyl-2H-1,4a-propanobenzofuro[2,3-c]pyridin-8-ol (5) [8][10], the ortho-f epoxy-
bridged isomer, rac-(3R,6aS,11aR)-1,3,4,5,6,11a-hexahydro-2-methyl-2H-3,6a-meth-
anobenzofuro[2,3-c]azocine-10-ol (6) [11], the ortho-d epoxy-bridged isomer, and
rac-(4R,6aR,11bR)-2,3,4,5,6,6a-hexahydro-3-methyl-1H-4,11b-methanobenzofuro[3,2-d]-
azocine-10-ol (7) [1], the para-a epoxy-bridged isomer (Fig. 2). We have also reported
an improved synthesis of the ortho-a epoxy-bridged phenylmorphan isomer [1]. We
now report the synthesis of rac-(3R,6aS,11aR)-1,3,4,5,6,11a-hexahydro-2-methyl-2H-
3,6a-methanobenzofuro[2,3-c]azocin-8-ol (3), the para-d epoxy-bridged phenylmor-
phan isomer.
Synthesis.
The synthesis of the target compound 3 consisted of two parts:
construction of the 5-phenylazabicyclo[3.3.1]nonane skeleton (Scheme 1) and subse-
quent formation of the epoxy bridge [11] (Scheme 2). In Scheme 1, treatment of alcohol
9 with acid, followed by alkylation and ring closure by the approach of Evans et al. [18]

Helvetica Chimica Acta Vol. 86 (2003)
487
Fig. 2. Structures of previously synthesizedepoxy-bridgedisomers
Scheme 1. Synthesis of the Key Intermediate, Enamine 11
a) BuLi. b) 99% H2SO4. c) Allyl bromide. d) N-Bromoacetamide.

488
Helvetica Chimica Acta Vol. 86 (2003)
Scheme 2. Closure of the Epoxy Bridge
H3 CO
H3CO
a)
N
CH3
N
CH3
Br
O
OCH 3
H3C
12
13
b)
H3CO
N
CH3
Br
O
H3C
14
d)
c)
HO
HO
N
CH3
N
CH3
Br
O
Br
OH
H3C
16
15
HO
CH3
N
O
N
CH3
O
HO
3
a) N-Bromoacetamide. b) NaBH3CN, HCl. c) Small excess BBr3. d) 10 equiv. BBr3, 0.5m KOH.
was envisaged to give the enamine 12, in which the CˆC bond afforded the necessary
functionality for the formation of the epoxy bridge.
The synthesis of 9 was straightforward, following the route earlier used for the
synthesis of the ortho-substituted compound [11]. Arylation of 1-methylpiperidin-4-

Helvetica Chimica Acta Vol. 86 (2003)
489
one with lithiated 1,4-dimethoxybenzene (8) gave, thus, the tertiary alcohol 9, which
afforded tetrahydropyridine 10 upon acid-catalyzed dehydration. A related tetrahy-
dropyridine has shown severe neurotoxic effects [19]. Metallation of 10 with BuLi at
À158, followed by alkylation of the anion with allyl bromide at À788, resulted in an
80% yield (by GC) of the terminal olefin 11. When the crude vinylic compound 11,
without further purification, was treated with HCOOH/H3PO4 for 50 h at room
temperature, 12 could be isolated in 52% yield (based on 10) as its oxalate salt.
Attempted purification of crude 11 by distillation (1 5 mm, head 210 2158, bath 235
2508) gave a complex mixture in which the desired compound 11 could not be detected.
However, after subjecting the complex mixture to chromatography and crystallization,
a 10% yield of 12 (isolated as the oxalate salt) was obtained, presumably formed via
thermal rearrangement of 11 during the distillation.
With compound 12 in hand, the stage could be set for the formation of the epoxy
bridge. Bromination of the enamine with N-bromoacetamide, followed by reduction of
the CˆC bond with NaBH3CN under acidic conditions afforded 14 as a single epimeric
compound (Scheme 2). The configuration could not be determined unambiguously
from the NMR spectrum, but from the result of the subsequent ring closure, it may be
concluded that the Br-atom was in the b-position. Treatment of 14 with a small excess
of Br3B at room temperature [20] did not bring about the desired didemethylation.
Instead, the mono-hydroxy compound 15 was obtained in good yield. From the fact that
this compound did not afford a ring-closed product upon treatment with t-BuOK in
THF [11], we concluded that the less-hindered group had been attacked. Heating a
solution of 14 with 10 equiv. of Br3B at 608 for 3 h gave rise to a more polar product,
presumably diphenol 16, which, upon workup with aqueous KOH, gave the desired
1
epoxy compound 3. The H-NMR spectrum showed a double doublet at d 4.62 ppm
(J ˆ 6.5 and 8 Hz), typical of a H-atom adjacent to an epoxy bridge. The structure of 3
was established unambiguously by single-crystal X-ray-analysis (Fig. 3).
Fig. 3. X-Ray crystal structure of rac-(3R,6aS,11aR)-1,3,4,5,6,11a-hexahydro-2-methyl-2H-3,6a-methanobenzo
furo[2,3-c]azocin-8-ol (3) (drawn from the experimentally determined coordinates with displacement ellipsoids
at the 20% probability level)

490
Helvetica Chimica Acta Vol. 86 (2003)
Results and Discussion.
The para-d isomer, rac-(3R,6aS,11aR)-1,3,4,5,6,11a-
hexahydro-2-methyl-2H-3,6a-methanobenzofuro[2,3-c]azocin-8-ol (3) was successfully
synthesized. As determined by an X-ray crystallographic study, the epoxy bridge,
restricting the phenyl-ring rotation, fixes the dihedral angle between the least-squares
planes through the phenyl ring and atoms N(2), C(3), C(11a), and C(6a) of the
piperidine ring (Fig. 3) at 43.0(1)8, and the torsion angle C(12)ÀC(6a)ÀC(6b)ÀC(10a)
at À95.0(2)8. A comparison of structural parameters to those in the ortho-d isomer
[11], which differs by the location of the OH substituent at C(10) vs. C(8), shows that
only a few of the many shared bond distances and angles differ by as much as three
standard deviations, and the comparable dihedral angle between the phenyl and
piperidine rings is 498 in the ortho-d isomer [11]. The relationship of the dihedral angle
noted above and the opioid receptor affinity and pharmacological properties of the
epoxy-bridged phenylmorphans will be reported in due course.
The authors (LMC, NIDDK) thank the National Institute on Drug Abuse, NIH, for partial financial support
of our research program. The X-ray crystallographic work was supported by the Office of Naval Research and by
the National Institute on Drug Abuse, NIH.
Experimental Part
General. M.p.: Thomas-Hoover melting-point apparatus; uncorrected. TLC: 250-m Analtech GHLF silica
gel plates with CHCl3/MeOH/conc. NH4OH 85 :15 :0.5 as the solvent system. Column chromatography (CC):
230 400 mesh silica gel 60 (Fluka 60738). GC: Hewlett-Packard5890 instrument with a flame-ionization
detector. IR Spectra: Beckman 4230 IR spectrophotometer. 1H-NMR Spectra: Varian XL-300 instrument, with
CDCl3 (TMS as internal standard). Chemical-ionization mass spectra (CIMS) were obtained using a Finnigan
1015 mass spectrometer. Electron-ionization mass spectra (EI-MS): VG-Micro Mass 7070F mass spectrometer.
Elemental analyses: Atlantic Microlabs, Inc., Norcross, GA; within Æ 0.4% of the theoretical values.
4-(2,5-Dimethoxyphenyl)-1-methylpiperidin-4-ol (9). A soln. of 1,4-dimethoxybenzene (8, 207.0 g, 1.50 mol)
in 1300 ml of dry Et2O was cooled in an ice bath and stirred under Ar while a 1.6m soln. of BuLi (775 ml,
1.24 mol) was added over a period of 20 min. The ice bath was removed, and the soln. was stirred for 24 h. The
resulting white suspension was cooled to 08, and a soln. of 1-methylpiperidin-4-one (140.1 g, 1.24 mol) in 150 ml
of dry Et2O was slowly added (20 min). The light yellow, clear soln. was stirred for 45 min. To the mixture was
added 400 ml of sat. aq. NaHCO3, the mixture was stirred for 20 min and filtered. The layers were separated, and
the org. phase was washed with aq. NaHCO3 (2 Â 200 ml) and H2O (2 Â 100 ml). The org. phase was partly
evaporated, and crystalline 9 (109.0 g, 0.43 mol, 38% yield based on amine) was obtained. The mother liquor
was extracted with 450 ml of 4n HCl. The aq. layer was separated, basified with 28% NH4OH, and extracted
with Et2O (3 Â 100 ml). The combined org. phase was washed with H2O (50 ml), dried (Na2SO4), and
evaporated until 50 ml of Et2O was left. A second crop of 9 (30.0 g, 0.12 mol, total yield 44%) was collected.
M.p. 103 1048. 1H-NMR (CDCl3): 2.35 (s, MeN); 3.77 (s, MeO); 3.87 (s, MeO); 6.74 6.91 (m, 3 arom. H). CI-
‡
‡
MS (NH3): 252 ([M ‡ H] ), 234 ([M ‡ H À H2O] ). Anal. calc. for C14H21NO3: C 66.91, H 8.42, N 5.57; found:
C 66.69, H 8.44, N 5.53.
4-(2,5-Dimethoxyphenyl)-1,2,3,6-tetrahydro-1-methylpyridine (10). Alcohol 9 (109.0 g, 0.43 mol) was
suspended in 125 ml of H2O, and conc. H2SO4 (60 ml) was cautiously added while stirring. The mixture was
heated to boiling and refluxed for 1 h, at which time complete conversion was observed on TLC. The soln. was
cooled to r.t., made alkaline (pH 10) with 40% NaOH, and extracted with Et2O (2 Â 100 ml). The combined org.
phase was washed with 0.5m KOH (50 ml) and H2O (50 ml), dried (Na2SO4), and evaporated to give 100.5 g
(0.4 mol, 100%) of 10 (CAUTION)3) as a light-yellow oil, pure enough for further use. For analysis, a sample
was converted to the HCl salt with ethanolic HCl and recrystallized from EtOH. M.p. 236 2378 (dec.) 1H-NMR
(CDCl3, free base): 2.38 (s, MeN); 3.07 (m, 2 H); 2.55 (m, 2 H); 2.62 (m, 2 H); 3.75 (s, MeOÀC(2'),
3
)
Studies have indicated that similar 1,2,3,6-tetrahydro-4-phenylpyridines are extremely potent neurotoxins.
Care should be taken in handling this material [19].

Helvetica Chimica Acta Vol. 86 (2003)
491
‡
MeOÀC(5')); 5.77 (s, HÀC(5)); 6.70 6.82 (m, 3 arom. H). CI-MS (NH3): 234 ([M ‡ H] ). Anal. calc. for.
C14H19NO2 ¥ HCl: C 62.33, H 7.47, N 5.19; found: C 62.32, H 7.64, N 5.21.
4-(2,5-Dimethoxyphenyl)-1,2,3,4-tetrahydro-1-methyl-4-(prop-2-enyl)pyridine (11). Compound 10 (20.3 g,
87 mmol) was dissolved in 150 ml of dry THFand cooled to À158 (ice-salt bath) under Ar. BuLi (80 ml of a 1.6m
soln. in hexane) was introduced by syringe, producing a dark-red soln. This was stirred for 20 min, cooled to
À788, and allyl bromide (15.5 g, 0.128 mol) in 20 ml of dry THF was added. After the addition, the yellow soln.
was warmed to r.t. H2O (50 ml) was slowly added, and the layers were separated. The aq. soln. was extracted
with Et2O (2 Â 100 ml), and the combined org. phase was washed with H2O (2 Â 100 ml). Drying (Na2SO4) and
evaporation gave 22.1 g of an orange oil, containing 80% of 11 according to GC. This material was used in the
next step without further purification. 1H-NMR (CDCl3, free base): 2.56 (s, MeN); 3.77 (s, MeO); 3.80 (s,
MeO); 4.61 (dd, J ˆ 2, 8, HÀC(3)); 4.86 4.98 (m, CHˆCH2); 5.46 5.52 (m, CHˆCH2); 5.95 (d, J ˆ 8,
‡
HÀC(2)); 6.67 7.00 (m, 3 arom. H). CI-MS (NH3): 274 ([M ‡ H] ).
rac-5-(2,5-Dimethoxyphenyl)-2-methyl-2-azabicyclo[3.3.1]non-3-ene Oxalate (12 ¥ Oxalate). Crude 11
(20.2 g) was dissolved in a mixture of 20 ml of anh. HCOOH and 20 ml of 85% H3PO4 at 08. The mixture
was stirred for 50 h at r.t. The mixture was diluted with 300 ml of H2O, cooled in ice, and made alkaline with 30%
KOH. Extraction with Et2O (3 Â 50 ml), washing with H2O (50 ml), drying (Na2SO4), and evaporation of the
solvent gave 18.1 g of a dark-brown oil. To a soln. of this oil in 75 ml of boiling i-PrOH, oxalic acid (6.6 g,
73 mmol) in 150 ml of hot i-PrOH was added. Crystallization began immediately, yielding 14.90 g (41.0 mmol,
52% based on 10) of 12 ¥ oxalate. An anal. sample was recrystallized from MeOH/acetone, M.p. 193 1968 (dec.,
ending in red melt at 208 2108); 1H-NMR (CDCl3, free base): 2.74 (s, MeN); 3.77 (s, MeO); 3.80 (s, MeO); 4.62
‡
(dd, J ˆ 7.8, 2, HÀC(4)); 6.05 (d, J ˆ 7.8, HÀC(3)), 6.66 6.92 (m, 3 arom. H). CI-MS (NH3): 273 (M ), 230.
Anal. calc. for C17H23NO2 ¥ C2H2O4: C 62.80, H 6.93, N 3.85; found: C 62.54, H 6.85, N 3.80.
rac-4-Bromo-5-(2,5-dimethoxyphenyl)-2-methyl-2-azabicyclo[3.3.1]non-3-ene (13). The oxalate of 12
(6.30 g, 17 mmol) was converted to the free base by partitioning between CHCl3 and 1m KOH to give 4.86 g
(17 mmol) of 12. This material was dissolved in 40 ml of dry THF, cooled to À788, and a soln. of N-
bromoacetamide (2.76 g, 20 mmol) in 20 ml of dry THF was added. After stirring for 20 min at À788, the
mixture was allowed to warm to r.t. The solvents were evaporated, and the oily residue was dissolved in 50 ml of
CHCl3 and washed with sat. NaHCO3 (2 Â 20 ml) and H2O (10 ml). The org. fraction was dried (Na2SO4) and
evaporated. Chromatography of the resulting oil over silica gel (1% MeOH in CHCl3) gave 2.83 g (8.0 mmol,
1
47%) of 13 as an orange oil, homogeneous in TLC. This material was used directly in the next step. H-NMR
(CDCl3): 2.75 (s, MeN); 3.77 (s, MeO); 3.80 (s, MeO); 6.38 (s, HÀC(3)); 6.73 6.87 (m, 3 arom. H).
rac-4-Bromo-5-(2,5-dimethoxyphenyl)-2-methyl-2-azabicyclo[3.3.1]nonane Oxalate (14 ¥ Oxalate). Com-
pound 13 (2.80 g, 8.0 mmol) was dissolved in 125 ml of MeOH, 3 ml of conc. HCl was added, followed by 0.61 g
(9.6 mmol) of NaBH3CN. The soln. was stirred for 20 min at r.t. Sat. NaHCO3 soln. (100 ml) and H2O (20 ml)
were added, and the mixture was extracted with 3 Â 50 ml of CH2Cl2. The combined org. phase was evaporated
to give 2.67 g of a red oily foam. This material was dissolved in 20 ml of MeCN and treated with a soln. of 0.75 g
(8.3 mmol) oxalic acid in 5 ml of MeCN. The mixture was heated to boiling and set aside. Crystallization started
immediately, giving 1.97 g (4.4 mol, 55%) of 14 ¥ oxalate, M.p. 178 1798 (with foaming). 1H-NMR (CDCl3, free
base): 2.47 (s, MeN); 3.21 (dd, J ˆ 7.2, 12, HÀC(3)); 3.43 (m, 2 H); 3.77 (s, MeO); 3.83 (s, MeO); 5.57 (dd, J ˆ
‡
‡
7.2, 11.2, HÀC(4)); 6.72 6.93 (m, 3 arom. H). CI-MS (NH3): 354, 356 ([M ‡ H] ), 274 ([M ‡ 1 À HBr] ).
Anal. calc. for C17H24BrNO2 ¥ C2H2O4: C 51.36, H 5.90, N 3.15; found: C 51.26, H 5.87, N 3.21.
rac-3-(4-Bromo-2-methyl-2-azabicyclo[3.3.1]non-5-yl)-4-methoxyphenol (15). Compound 14 (100 mg,
0.28 mmol) was dissolved in 3 ml of pentene-stabilized CHCl3, and 1.6 ml of a 1m soln. of Br3B in CH2Cl2
(1.6 mmol) was added by syringe. After stirring for 20 min at r.t., the mixture was poured into a mixture of ice
(10 g) and 3 ml of conc. NH4OH. The layers were separated, and the aq. layer was extracted with CHCl3 (2 Â
10 ml). The combined org. phase was washed with H2O (10 ml), dried (Na2SO4), and evaporated to give 60 mg
1
(0.22 mmol, 77%) of pure 15. Crystals from EtOH. M.p. 158 1598. H-NMR (CDCl3): 2.48 (s, MeN); 3.82 (s,
‡
‡
MeO); 5.55 (m, HÀC(4)); 6.67 6.85 (m, 3 arom. H). CI-MS (NH3): 340, 342 ([M‡ H] ), 260 ([M‡ HÀ HBr] ).
Anal. calc. for C16H22BrNO2 ¥ 0.75 H2O: C 54.32, H 6.70, N 3.96; found: C 54.44, H 6.40, N 3.95.
rac-(3R,6aS,11aR)-1,3,4,5,6,11a-Hexahydro-2-methyl-2H-3,6a-methanobenzofuro[2,3-c]azocin-8-ol Hy-
drochloride (3 ¥ HCl) and Hydrobromide (3 ¥ HBr). The oxalate salt of compound 14 (1.30 g, 2.97 mmol) was
converted to the free base (1m KOH, CHCl3) to give 1.05 g (2.97 mmol, 100%) of an oil. The oil was dissolved in
15 ml of pentene-stabilized CHCl3, and, under N2, Br3B (2.8 ml, 30 mmol) was added by syringe over a period of
5 min. The mixture was stirred at r.t. for 30 min, after which TLC showed complete conversion of the starting
material (Rf 0.90) to the phenol 15 (Rf 0.62). The soln. was heated to reflux (oil-bath temp. 708) for 3 h, after
which time TLC showed the complete conversion of 15 into a more polar compound (presumably 16, Rf 0.38).

492
Helvetica Chimica Acta Vol. 86 (2003)
After cooling to r.t., the mixture was slowly added to a vigorously stirred soln. of KOH (8.0 g, 142 mmol) in a
1:1 mixture of H2O and MeOH (250 ml), held at 08 under N2. The mixture was stirred for 20 min, after which
TLC showed clean conversion to a less polar material (Rf 0.55). NH4Cl (5 g) was added, and the layers were
separated. The aq. phase was extracted with CHCl3/EtOH 4 :1 (2 Â 25 ml). The combined org. phase was dried
(Na2SO4) and evaporated to give 0.6 g of 3 as a light-brown foam. Crude 3 was dissolved in 15 ml of AcOEt, and
5 ml of ethanolic HCl was added. More EtOH (5 ml) was added, and the precipitate was dissolved by heating.
Upon cooling to r.t. and scratching, 3 ¥ HCl crystallized (0.50 g, 60%). Recrystallization from EtOH gave
irregular hygroscopic crystals. M.p. 234 2358. The HBr salt of 3 was crystallized from MeOH. M.p. 248 2508.
1H-NMR (CDCl3, free base): 1.18 1.36 (m, 2 H); 1.42 (dd, J ˆ 13, 1.9, 1 H); 1.58 2.00 (m, 6 H); 2.04 (d, J ˆ 13,
1 H); 2.2 (br. s, OH); 2.43 (s, MeN); 2.74 (br. s, 1 H); 2.80 (dd, J ˆ 4.5, 8, HÀC(1)); 3.06 (dd, J ˆ 6.5, 11.3,
‡
HÀC(1')); 4.62 (dd, J ˆ 8, 6.5, 1 H); 6.55 (dd, J ˆ 6.2, 2.6, 1 H); 6.58 (m, 2 H). CI-MS (NH3): 246 ([M ‡ H] ).
Anal. calc. for C15H19NO2 ¥ HCl ¥ 2 H2O:C 56.69, H 7.61, N 4.41; found: C 56.25, H 7.64, N 4.40. Anal. calc. for
C15H19NO2 ¥ HBr: C 55.23, H 6.18, N 4.29; found: C 55.00, H 6.26, N 4.29.
X-Ray Analysis4) of 3. For X-ray data, see the Table. Data were collected at r.t. with CuKa radiation on an
automated Bruker P4 diffractometer equipped with a monochromator in the incident beam. The crystal
remained stable during data collection. Face-indexed numerical absorption (minimum and maximum
transmission 0.040 and 0.56). The structure was solved by direct methods and refined by full-matrix least-
squares on F2 values using the SHELXTLplus system of programs [21]. Parameters (179) refined included
atomic coordinates and displacement ellipsoids for all non-H atoms. The H atoms on C-atoms were included
Table. Crystal Data andStrucutre Refinement of 3.
‡
Empirical formula
Formula weight
Temp. [K]
C15H20NO2 ¥ BrÀ
326.23
293(2)
Wavelength
1.54178
Crystal system
Space group
Monoclinic
P21/c
Unit cell dimensions
a ˆ 10.457(2) ä
b ˆ 10.541(3) ä
c ˆ 13.128(2) ä
1431.0(5) ä3
4
a ˆ 908.
b ˆ 98.55(2)8.
g ˆ 908.
Volume
Z
Density (calc.)
Absorption coefficient
F(000)
1.514 Mg/m3
3.897 mmÀ1
672
Crystal size
q Range for data collection
Index ranges
0.18 Â 0.18 Â 0.44 mm
4.28 to 57.348.
À 11 ꢀ h ꢀ 11, 0 ꢀ k ꢀ 11, 0 ꢀ l ꢀ 14
2131
Reflections collected
Independent reflections
Observed reflections (I > 2s(I))
Completeness to q ˆ 57.348
Refinement method
Data / restraints / parameters
Goodness-of-fit on F2
Final R indices (I > 2s(I))
R indices (all data)
Extinction coefficient
Largest diff. peak and hole
1956 (Rint ˆ 0.0145)
1901
99.9%
Full-matrix least-squares on F2
1956 / 0 / 179
0.961
R1 ˆ 0.0265, wR2 ˆ 0.0781
R1 ˆ 0.0272, wR2 ˆ 0.0788
0.0023(2)
0.251 and À0.395 e.äÀ3
4
)
Coordinates have been deposited with the Cambridge Crystallographic Data Centre as deposition No.
CCDC 189625. Copies of the data can be obtained, free of charge, on application to the CCDC, 12 Union
Road, Cambridge CB2 1EZ UK (deposit@CCDC.cam.ac.uk).

Helvetica Chimica Acta Vol. 86 (2003)
493
with a riding model (coordinate shifts of C applied to H-atoms with CÀH distance set at 0.96 ä). Coordinates
only were refined for OH H-atoms and H-atoms on N-atoms.
REFERENCES
[1] K. Yamada, J. L. Flippen-Anderson, A. E. Jacobson, K. C. Rice, Synthesis-Stuttgart 2002, 2359.
[2] E. L. May, Science 1973, 181, 407.
[3] E. L. May, J. G. Murphy, J. Org. Chem. 1955, 20, 1197; H. Ong, T. Oh-ishi, E. L. May, J. Med. Chem. 1974,
17, 133; H. Awaya, E. L. May, A. E. Jacobson, M. D. Aceto, J. Pharm. Sci. 1984, 73, 1867.
[4] H. Awaya, E. L. May, M. D. Aceto, H. Merz, M. E. Rogers, L. S. Harris, J. Med. Chem. 1984, 27, 536.
[5] E. L. May, M. Takeda, J. Med. Chem. 1970, 13, 805.
[6] M. D. Aceto, L. S. Harris, E. L. May, in −Problems of Drug Dependence 1982×, Ed. L. S. Harris, NIDA
Research Monograph 43, Washington, D.C., 1983, p. 399.
[7] M. Froimowitz, W. Pangborn, V. Cody, Chirality 1992, 4.
[8] T. R. Burke Jr., A. E. Jacobson, K. C. Rice, B. A. Weissman, J. V. Silverton, in −Problems of Drug
Dependence 1983×, Ed. H. E. Harris, NIDA Research Monograph 49, Washington DC, 1984, Vol. 49, p. 109.
[9] T. R. Burke Jr., A. E. Jacobson, K. C. Rice, J. V. Silverton, J. Org. Chem. 1984, 49, 1051.
[10] T. R. Burke Jr., A. E. Jacobson, K. C. Rice, J. V. Silverton, J. Org. Chem. 1984, 49, 2508.
[11] T. R. Burke Jr., A. E. Jacobson, K. C. Rice, B. A. Weissman, H. C. Huang, J. V. Silverton, J. Med. Chem.
1986, 29, 748.
[12] A. J. Hutchinson, R. de Jesus, M. Williams, J. P. Simke, R. F. Neale, R. H. Jackson, F. Ambrose, B. J. Barbaz,
M. A. Sills, J. Med. Chem. 1989, 32, 2221.
[13] A. Hashimoto, A. Coop, R. B. Rothman, C. Dersch, H. Xu, R. Horel, C. George, A. E. Jacobson, K. C.
Rice, in −Problems of Drug Dependence, 1999×, Ed. L. S. Harris, National Institute on Drug Abuse
Research Monograph 180, Washington DC, 2000, p. 250; A. Hashimoto, R. B. Rothman, C. Dersch, R.
Horel, A. E. Jacobson, K. C. Rice, Drug Alcohol Dependence 2000, 60, S86; A. Hashimoto, A. E. Jacobson,
R. B. Rothman, C. Dersch, C. George, J. L. Flippen-Anderson, K. C. Rice, Bioorg. Med. Chem. 2002, 10,
3319.
[14] J. B. Thomas, K. M. Gigstad, S. E. Fix, J. P. Burgess, J. B. Cooper, S. W. Mascarella, B. E. Cantrell, D. M.
Zimmerman, F. I. Carroll, Tetrahedron Lett. 1999, 40, 403; J. B. Thomas, X. Zheng, L. E. Brieaddy, J. P.
Burgess, S. W. Mascarella, S. E. Fix, B. E. Cantrell, D. M. Zimmerman, F. I. Carroll, Tetrahedron Lett. 1998,
39, 7001.
[15] J. B. Thomas, R. N. Atkinson, R. B. Rothman, J. P. Burgess, S. W. Mascarella, C. M. Dersch, H. Xu, F. I.
Carroll, Bioorg. Med. Chem. Lett. 2000, 10, 1281.
[16] D. M. Zimmerman, R. Nickander, J. S. Horng, D. T. Wong, Nature 1978, 275, 332.
[17] J. B. Thomas, X. L. Zheng, S. W. Mascarella, R. B. Rothman, C. M. Dersch, J. S. Partilla, J. L. Flippen-
Anderson, C. F. George, B. E. Cantrell, D. M. Zimmerman, F. I. Carroll, J. Med. Chem. 1998, 41, 4143.
[18] D. A. Evans, C. H. Mitch, R. C. Thomas, D. M. Zimmerman, R. L. Robbey, J. Am. Chem. Soc. 1980, 102,
5955.
[19] R. S. Burns, C. C. Chiueh, S. P. Markey, M. H. Ebert, D. M. Jacobowitz, I. J. Kopin, Proc. Natl. Acad. Sci.
U.S.A. 1983, 80, 4546; J. W. Langston, P. Ballard, J. W. Tertrud, I. Irwin, Science 1983, 219, 979.
[20] K. C. Rice, J. Med. Chem. 1977, 20, 164.
[21] G. M. Sheldrick, Version 5.03, Bruker Analytical X-Ray Instruments, Madison, WI, 1997.
ReceivedSeptember 19, 2002