Microbiological Oxygenation of Bridgehead Azabicycloalkanes

Article abstract of DOI:10.1021/jo962192f

A series of N-substituted bridgehead azabicycloalkanes has been prepared and examined as substrates for microbiological oxygenation using the fungi Beauveria bassiana, Rhizopus nigricans, Aspergillus ochraceus, and Rhizopus arrhizus. Oxygenation using B. bassiana of N-tosyl-7- azabicyclo[2.2.1]heptane gave N-[p-(hydroxymethyl)benzenesulfonyl]-7-azabicyclo[2.2.1]heptane (56% yield), of N-(phenyloxycarbonyl)-7-azabicyclo[2.2.1]heptane gave the 2-endo-ol (56% yield, 51% ee), of N-BOC-7-azabicyclo[2.2.1]heptane gave the 2-endo-ol (10% yield), of N-Cbz-7-azabicyclo- [2.2.1]heptane gave the 2-endo-ol (28%), of N-(phenyloxycarbonyl)-8-azabicyclo[3.2.1]octane gave the 3-endo-ol, and of N-(phenyloxycarbonyl)-9-azabicyclo[3.3.1]nonane gave the 3-exo-ol (30%) and 3-one (16%). Oxygenation using R. nigricans of N-BOC-7-azabicyclo[2.2.1]heptane gave the 2-endo-ol (62% yield, 28% ee) and the 2-exo-ol (27% yield, 42% ee). Oxidation of the N-BOC-7-azabicyclo- [2.2.1]heptan-2-ols gives the 2-ketone, a synthetic intermediate useful for conversion to the natural product, epibatidine. Oxygenation of N-(phenyloxycarbonyl)-7-azabicyclo[2.2.1]heptane using R. arrhizus gives the 2-endo-ol (5% yield, 31% ee) and the 2-exo-ol (18% yield, 22% ee). Oxygenation of N-(phenyloxycarbonyl)-8-azabicyclo[3.2.1]octane using A. ochraceous gives the 3-endo-ol (36%) and the 3-one (4%).

Full text of DOI:10.1021/jo962192f

2244
J . Org. Chem. 1997, 62, 2244-2251
Micr obiologica l Oxygen a tion of Br id geh ea d Aza bicycloa lk a n es
Charles R. Davis,*,1 Roy A. J ohnson,* J oyce I. Cialdella, Walter F. Liggett,
Stephen A. Mizsak, and Vincent P. Marshall
Research Laboratories, Pharmacia & Upjohn, Inc., Kalamazoo, Michigan 49001
Received November 25, 1996X
A series of N-substituted bridgehead azabicycloalkanes has been prepared and examined as
substrates for microbiological oxygenation using the fungi Beauveria bassiana, Rhizopus nigricans,
Aspergillus ochraceus, and Rhizopus arrhizus. Oxygenation using B. bassiana of N-tosyl-7-
azabicyclo[2.2.1]heptane gave N-[p-(hydroxymethyl)benzenesulfonyl]-7-azabicyclo[2.2.1]heptane
(56% yield), of N-(phenyloxycarbonyl)-7-azabicyclo[2.2.1]heptane gave the 2-endo-ol (56% yield, 51%
ee), of N-BOC-7-azabicyclo[2.2.1]heptane gave the 2-endo-ol (10% yield), of N-Cbz-7-azabicyclo-
[2.2.1]heptane gave the 2-endo-ol (28%), of N-(phenyloxycarbonyl)-8-azabicyclo[3.2.1]octane gave
the 3-endo-ol, and of N-(phenyloxycarbonyl)-9-azabicyclo[3.3.1]nonane gave the 3-exo-ol (30%) and
3-one (16%). Oxygenation using R. nigricans of N-BOC-7-azabicyclo[2.2.1]heptane gave the 2-endo-
ol (62% yield, 28% ee) and the 2-exo-ol (27% yield, 42% ee). Oxidation of the N-BOC-7-azabicyclo-
[2.2.1]heptan-2-ols gives the 2-ketone, a synthetic intermediate useful for conversion to the natural
product, epibatidine. Oxygenation of N-(phenyloxycarbonyl)-7-azabicyclo[2.2.1]heptane using R.
arrhizus gives the 2-endo-ol (5% yield, 31% ee) and the 2-exo-ol (18% yield, 22% ee). Oxygenation
of N-(phenyloxycarbonyl)-8-azabicyclo[3.2.1]octane using A. ochraceous gives the 3-endo-ol (36%)
and the 3-one (4%).
In tr od u ction
enantiomeric form of epibatidine.5 Once both enanti-
omers of the natural product became available, pharma-
cological evaluations have shown that both have anal-
gesic properties and are of approximately equal potency.6
The use of microorganisms for the selective oxygen-
ation of organic molecules is well documented; in par-
ticular, microbial transformations represent a powerful
tool for oxygenation of unactivated hydrocarbon sites.7
The list of substrates for biotransformation ranges from
steroids and arenes to a wide array of cyclic, bicyclic,
polycyclic, and acylic N-acylalkanes.8-10 Our laboratories
(-)-Epibatidine (1) is a naturally occurring 7-azabicyclo-
[2.2.1]heptane first isolated from the skin of the poison-
ous frog, Epipedobates tricolor.2 (-)-Epibatidine is a very
potent analgesic, but this desirable activity is accompa-
nied by serious toxicity, which has heightened interest
in the synthesis of analogs that may be free of side effects.
This, together with the lack of abundance of the natural
product, have led to intensive efforts aimed at the total
synthesis of the compound. Numerous syntheses both
of racemic and of the natural and unnatural enantiomers
of epibatidine have been reported.3 Fletcher and co-
workers obtained (+)- and (-)-epibatidine through reso-
lution and elaboration of (()-alcohol 2.4 In an enan-
have considerable experience in performing microbiologi-
cal oxygenation reactions including the oxygenation of a
variety of N-acyl heterobicyclic alkanes.11 However, the
bridgehead azabicycloalkane series of 7-azabicyclo[2.2.1]-
heptane,12 8-azabicyclo[3.2.1]octane, and 9-azabicyclo-
[3.3.1]nonane has not been the subject of such oxygen-
ation studies. We envisioned that selective oxygenation
tiospecific synthesis reported by Rapoport and co-
workers, diastereomeric mixtures formed in the course
of the synthesis did not require separation but instead
converged on a single enantiomer, either (+)- or (-)-3,
which in turn could be converted to the respective
(5) Herna´ndez, A.; Marcos, M.; Rapoport, H. J . Org. Chem. 1995,
60, 2683.
(6) Bai, D.; Xu, R.; Chu, G.; Zhu, X. J . Org. Chem. 1996, 61, 4600.
(7) J ohnson, R. A. In Oxidation in Organic Chemistry; Trahanovsky,
W. S., Ed.; Academic Press: New York, 1978; Part C, p 131.
(8) Holland, H. L. Organic Synthesis With Oxidative Enzymes;
VCH: New York, NY, 1992.
(9) Davies, H. G.; Green, R. H.; Kelley, D. R.; Roberts, S. M.
Biotransformations in Preparative Organic Chemistry: The Use of
Isolated Enzymes and Whole Cell Systems; Harcourt Brace J ovanov-
ich: New York, NY, 1989.
(10) Vigne, B.; Archelas, A.; Furstoss, R. Tetrahedron 1991, 47, 1447
and references cited therein.
X Abstract published in Advance ACS Abstracts, March 15, 1997.
(1) Postdoctoral research fellow at Pharmacia & Upjohn, J anuary
1, 1995-March 1, 1996. Present address: Department of Chemistry,
Augustana College, Rock Island, IL 61201.
(2) Spande, T. F.; Garraffo, M. H.; Edwards, M. W.; Yeh, H. J . C.;
Pannel, L.; Daly, J . W. J . Am. Chem. Soc. 1992, 114, 3475.
(3) Early synthesis: Broka, C. A. Tetrahedron Lett. 1993, 34, 3251.
A recent listing of syntheses may be found in: Zhang, C.; Trudell, M.
L. J . Org. Chem. 1996, 61, 7189.
(11) J ohnson, R. A.; Herr, M. E.; Murray, H. C.; Reineke, L. M.;
Fonken, G. S. J . Org. Chem. 1968, 33, 3195.
(12) The chemistry of this ring system has been reviewed: Chen,
Z.; Trudell, M. L. Chem. Rev. 1996, 96, 1179.
(4) Fletcher, S. R.; Baker, R.; Chambers, M. S.; Herbert, R. H.;
Hobbs, S. C.; Thomas, S. R; Verrier, H. M.; Watt, A. P.; Ball, R. G. J .
Org. Chem. 1994, 59, 1771.
S0022-3263(96)02192-5 CCC: $14.00 © 1997 American Chemical Society

Oxygenation of Bridgehead Azabicycloalkanes
J . Org. Chem., Vol. 62, No. 7, 1997 2245
of N-acyl-7-azabicylo[2.2.1]heptanes 4 might lead to
intermediates 5 or 6 useful for the synthesis of epibati-
dine and analogs thereof.
others and from these laboratories.7 Initially, substrates
8, 9, and 11 were each bioconverted using B. bassiana,
R. nigricans, and A. ochraceous. Workup of fermentation
samples taken periodically followed by analysis of crude
extracts by GCMS allowed us to detect possible oxygen-
ated products and to estimate the extent of conversion
of starting materials. The crude extracts were diluted
to a concentration theoretically equivalent to that of
prepared standards. Direct comparison of integration
(GCMS) allowed for approximation of conversion to
oxygenated products and overall recovery. From these
results, we were able to choose conditions for scale up.
Large-scale fermentations were monitored starting at the
3 day point and every 12-24 h thereafter. Large-scale
fermentation times were 3-6.5 days in length.
Our objective in this work, therefore, has been to obtain
stereoselective oxygenation of the 7-azabicyclo[2.2.1]-
heptane moiety and to study the outcome of these
oxygenations as influenced by changes in both the
N-substituent and the microorganism. We planned to
extend our studies to the oxygenation of 8-azabicyclo-
[3.2.1]octane and 9-azabicyclo[3.3.1]nonane substrates.
For the microorganisms employed in these studies, the
stereochemical information contained in the products
should contribute to a better understanding of both the
regio- and stereochemistry of substrate oxygenation.
Described herein are the microbial hydroxylations of a
series of N-acyl-7-azabicyclo[2.2.1]heptanes, of 8-(phen-
yloxycarbonyl)-8-azabicyclo[3.2.1]octane, and of 9-(phe-
nyloxycarbonyl)-9-azabicyclo[3.3.1]nonane using the fungi
Beauveria bassiana, Rhizopus nigricans, Aspergillus
ochraceus, and Rhizopus arrhizus as the primary organ-
isms.
We were surprised to find that in the bioconversion of
8 with B. bassiana, the product (13) was exclusively the
result of hydroxylation of the p-phenylmethyl group in
preference to oxygenation of the bicyclic ring carbons. The
1
site of hydroxylation was easily determined from the H
NMR spectrum of the product 13. The hydroxymethyl
Resu lts a n d Discu ssion
N-Acyl-7-a za bicyclo[2.2.1]h ep ta n es. A series of
N-acyl-7-azabicyclo[2.2.1]heptane substrates was pre-
pared by acylation of the hydrochloride salt 713 under
Schotten-Baumann conditions. The characterization of
product 13 was also the only product formed, as detected
by GCMS, in small-scale transformations of 8 with the
microorganisms R. nigricans and A. ochraceous. In each
case, the GCMS fragmentation pattern and retention
time of the product were identical to that observed for
13. Conversion of 8 to 13 using R. nigricans was slightly
higher than was observed for B. bassiana, but only a
trace of 13 was detected in a fermentation using A.
ochraceus.
The other N-acyl-7-azabicyclo[2.2.1]heptane substrates
(9-12) yielded products in which hydroxylation occurred
only on the bicycloalkane moiety. We first examined
bioconversion of phenyloxycarbonyl derivative 9. Hy-
droxylation of 9 using B. bassiana gave a single product
in an isolated yield of 46% that was characterized as the
endo-alcohol 14 (see Table 1). Substrate 9 is achiral, but
substitution of any of the methylene groups will introduce
chirality into the molecule. Product 14 has [R]D +5.2°,
indicating that a degree of asymmetric induction occurred
during the oxygenation process. Note that the structures
used to illustrate this discussion are drawn to represent
the absolute configuration of the enantiomer obtained in
excess from the oxygenation process. In a small-scale
bioconversion of 9 with R. nigricans, a 20% yield (deter-
mined by GCMS) of 14 was observed, while no product
was detected in a fermentation using A. ochraceus.
The assignment of configuration to the hydroxyl group
in 14 is based on NMR data accumulated at 50 °C;
sharper signals were observed at this temperature, which
allowed for a more detailed analysis of coupling con-
stants. A 13C DEPT spectrum exhibited three methylene
carbon signals, six tertiary carbon signals, and two
quaternary carbon signals, consistent with assignment
of hydroxylation to a secondary carbon. 1H NMR cou-
plings observed for H-2 were consistent with the vicinal
Karplus angles required by the endo-hydroxy stereo-
derivatives 8-12 was straightforward, although we
observe sets of signals for each of two rotamers in some
NMR spectra as a consequence of restricted rotation
about the N-CO bond. This phenomenon was described
for the N-acetyl derivative of epibatidine.2 We find that
the number of chemically nonequivalent carbons in the
13C NMR spectra, at room temperature, is dependent on
the nature of the N-substituent. In the 13C NMR
spectrum of benzamide 12, four distinct signals are
observed for the bicycloalkane methylene carbons. In
contrast, only two sharp signals are observed for the
bicycloalkane methylene carbons in derivatives 8 and 11,
whereas two broader signals are observed for 9 and 10.
NMR evidence for restricted rotation has frequently been
reported for other amides as well.14
Exploratory bioconversions were carried out in shake
flasks on 10-20 mg of substrate in 100 mL of nutrient
media. Large-scale conversions were carried out on 0.5-
2.0 g of substrate in 10 L of media either in a 10 L
fermentation tank or in multiple shake flasks. Microor-
ganisms were selected from inspection of the considerable
wealth of microbial transformations reported both from
(13) Hassner, A.; Belostotskii, A. M. Tetrahedron Lett. 1995, 36,
1709.
(14) Iida, H.; Watanabe, Y.; Kibayashi, C. J . Org. Chem. 1985, 50,
1818.

2246 J . Org. Chem., Vol. 62, No. 7, 1997
Davis et al.
Ta ble 1. Resu lts of th e Oxygen a tion of N-Acyl-7-a za bicyclo[2.2.1]h ep ta n es Usin g Sever a l Micr oor ga n ism s
compd
microorganism
compd
yield (%)
abs confign
eea (%)
compd
yield (%)
abs confign
eea (%)
9
9
B. bassiana
R. nigricans
B. bassiana
R. nigricans
B. bassiana
R. nigricans
R. arrhizus
14
14
15
15
16
16
16
46
20
10
62
28
14
5
(1S)
ndb
51
nd
26
28
5
10
10
11
11
11
(1S)
(1S)
(1R)
(1S)
(1S)
2
2
17
17
17
0
27
0
8
18
(1S)
42
36
31
(1S)
(1S)
57
22
a
b
Enantiomeric excess (ee) determined by 1H and 19F NMR of Mosher esters (see Experimental Section). Not determined.
chemistry (see illustration of 14). The observed couplings
uct. This absolute configuration is opposite that found
in natural (-)-epibatidine.4
We then explored the stereoselectivity of the oxygen-
ation process as a consequence of changes in the N-
substituent (see Table 1). In small-scale bioconversions
of the BOC derivative 10, major and minor hydroxylated
products were detected with either B. bassiana and R.
nigricans but not with A. ochraceous. The same two
products were formed in either fermentation judging from
the GCMS data. When the bioconversion of 10 with B.
bassiana was scaled up in a 10 L tank, the minor product
was detected in trace amounts and only the endo alcohol
15 was isolated. GCMS analysis prior to workup sug-
gested a high yield of 15, but we experienced difficulties
in isolation of the pure product and obtained this
compound in only 10% yield. The NMR data for the
product are identical to that reported in the literature
for 15. Following oxidation to ketone 3, correlation of
optical rotation to the literature data5 for enantiomeri-
cally pure 3 gave an ee of 26% and 1S absolute config-
uration. The 26% ee for 15 was confirmed by NMR
analysis of Mosher esters 19a and 19b.
A large-scale bioconversion of BOC-derivative 10 using
R. nigricans afforded both endo alcohol 15 and exo alcohol
2. The isomers were separable by SiO2 chromatography
and were isolated in an excellent total yield of 89%.
Enantiomerically pure 2 has been reported,4 and com-
parison of NMR and optical rotation data allowed as-
signment of the absolute configuration and ee as shown
in Table 1. Following oxidation of 15 to ketone 3,
correlation of the optical rotation with literature data5
gave an ee of 28% that again was confirmed by analysis
of Mosher esters 19a and 19b derived from 15.
were confirmed by a COSY experiment, including the
coupling between H-2 and H-6exo. A correlation between
C-2 (δ 70.6) and H-2 was also observed in a HETCOR
experiment. Similar H-H couplings are reported for
other endo N-substituted 7-azabicyclo[2.2.1]heptan-2-ols.4
The configuration of the hydroxyl group was confirmed
by conversion of 14 to the N-BOC derivative 15, which
has been reported as the racemate4 and also is obtained
from bioconversions described below using N-BOC sub-
strate 10. Hydrolysis of the phenyloxycarbonyl group of
14 was achieved with aqueous base using 18-crown-6 as
a phase-transfer catalyst. Consumption of starting ma-
terial was complete after 18 h of reflux, and after reaction
with BOC anhydride, an 85% yield of 15 was obtained.
When we did the hydrolysis of 14 using tert-butyl
ammonium iodide as the phase-transfer catalyst,15 the
reaction required 48 h and a lower yield of 15 was
isolated.
The enantiomeric excess of 14 was measured in two
ways. First, the Mosher esters 18a and 18b were
prepared, and the ratio of diastereoisomers was quanti-
tated by both 1H and 19F NMR. Second, the alcohol 15
derived above from 14 was oxidized via a Swern oxidation
to ketone 3 allowing correlation of the optical rotation to
that reported5 for the optically pure 3. Both approaches
gave ee values of 51% that can be assigned to 14.
Comparison of the sign of the optical rotations of the
ketones from the two sources allows assignment of 1S
as the absolute configuration of the bioconversion prod-
We next examined the N-(benzyloxycarbonyl) (Cbz)
derivative 11 as a potential substrate. Hydroxylated
products were detected in exploratory bioconversions of
11 using B. bassiana and R. nigricans but not with A.
ochraceus. A large-scale bioconversion using B. bassiana
gave an alcohol to which we have assigned structure 16
on the basis of an 1H NMR pattern similar to those
observed for 14 and 15. To confirm the structure, the
Cbz group was removed by transfer hydrogenation fol-
lowed by derivatization of the crude amine with tert-butyl
dicarbonate to give 15. 1H and 13C NMR data obtained
(15) Nelsen, S. F.; Hollinsed, W. C.; Kessel, C. R.; Calabrese, J . C.
J . Am. Chem. Soc. 1978, 100, 7876.

Oxygenation of Bridgehead Azabicycloalkanes
J . Org. Chem., Vol. 62, No. 7, 1997 2247
on this derivative correlated with NMR data acquired on
15 obtained in prior fermentations. A minor product,
identical to 17 (described below) in GCMS retention time
and fragmentation pattern, was detected in the explor-
atory bioconversion of 11 using B. bassiana. This minor
product was formed to a much lesser extent in the large-
scale fermentation, and only the endo isomer 16 was
isolated. A large-scale bioconversion of 11 using R.
nigricans produced both endo alcohol 16 and exo alcohol
17. The enantiomeric excess of endo alcohol 16 was
determined by analysis of Mosher esters 20a and 20b,
while characterization of exo alcohol 17 was carried out
by comparison of NMR and optical rotation data to that
obtained for 17 as described below.
first suggests that the seven carbons of the bicyclic ring
system are nonequivalent and that the hydroxyl group
is in an asymmetric location on the molecule. Inspection
of the spectrum, however, shows three closely paired sets
of signals, raising the possibility that equivalent carbon
atoms are giving two signals each because of unequal
magnetic fields generated by slow rotation of the acyl
group about the carbonyl-nitrogen bond. Indeed, signals
for C-1 and C-5, C-2 and C-4, and for C-6 and C-7
coalesced into broad singlets upon acquisition of the 13C
NMR spectrum at 50 °C. Consequently, the hydroxyl
group in 23 must be symmetrically placed on the mol-
ecule, and the C-3 position is the only logical choice for
this substitution.
An effort was then made to identify other organisms
capable of biotransformation with this class of substrates.
A limited screen of fungi (ca. 50 microorganisms) was
carried out on substrate 11, and several were identified
as producing 16 and 17 with R. arrhizus being the most
productive. A large-scale bioconversion of 11 using R.
arrhizus yielded endo alcohol 16 and exo alcohol 17 and
was the only experiment in which the exo alcohol
predominated (see Table 1). After chromatographic
separation, the structure and absolute configuration of
exo alcohol 17 were determined by conversion to the
N-BOC derivative 2 and comparison of NMR and optical
rotation data to that reported for enantiomerically pure
2.4
The melting point for our sample of 23 (151-153 °C)
was identical to the literature value (151-153 °C) for the
3-endo alcohol,16 but a mp for the epimeric alcohol has
not been reported. We confirmed the configuration of the
hydroxyl group in 23 by repeating the literature conver-
sion of tropine 25 to the Cbz derivative 23 as reported.16
1H and 13C NMR and melting point data obtained on the
authentic sample correlated with data obtained on the
23 we obtained from biotransformation.
Finally, in the case of benzoyl derivative 12, hydroxy-
lated products were detected (GCMS) on an exploratory
scale in bioconversions using B. bassiana and R. nigri-
cans, but these products were not isolated and character-
ized due to low conversion (5-10%) of starting material.
8-(P h e n y lo x y c a r b o n y l)-8-a z a b i c y c lo [3.2.1]-
oct a n e a n d 9-(P h en yloxyca r b on yl)-9-a za b icyclo-
[3.3.1]n on a n e. We have extended our biotransformation
studies to the 8-azabicyclo[3.2.1]octane and 9-azabicyclo-
[3.3.1]nonane systems. Phenyloxycarbonyl derivatives 21
and 22 were chosen as model substrates and were readily
prepared by reaction of the N-methyl precursors, tropane
and deoxopsuedopelletierine,15 with phenyl chlorofor-
mate.
Bioconversion of 21 using A. ochraceous gave an alcohol
(23, 36%) and a ketone (24, 4%). The 13C NMR spectrum
Bioconversion of 22 using A. ochraceus also gave an
alcohol (26, 30%) and a ketone (27, 16%). Assignment
of the oxygen substituent to the C-3 position in both
compounds could be made on the basis of analysis of the
1H and 13C NMR spectral data (both molecules give
spectra having only five nonaromatic carbon signals,
indicating symmetrical placement of the substituents).
The configuration of the hydroxyl group in 26 was
determined by conversion of carbamate 26 to benzamide
28.17 The melting point of 28 (153-154 °C) compares
well with the reported melting point (152-153 °C) for
the exo alcohol and differs from the reported melting
point (116-117 °C) for the endo alcohol.17
We were surprised that hydroxylation of 21 and 22 had
given endo and exo alcohols, respectively, since the
structures of the two substrates are quite similar. One
clear difference between the experiments is that they
used different microorganisms for the bioconversions of
the two compounds. We, therefore, carried out a small-
scale bioconversion of 21 using B. bassiana and examined
it by GCMS. The analytical results showed that alcohol
of the alcohol 23 exhibited seven signals for saturated
carbon atoms in the room-temperature spectrum (see
Experimental Section). The presence of seven signals at
(16) Leete, E. Phytochemistry 1972, 11, 1713.
(17) Alder, K.; Dortmann, H. A. Chem. Ber. 1953, 86, 1544.

2248 J . Org. Chem., Vol. 62, No. 7, 1997
Davis et al.
organic phases were dried (Na2SO4) and concentrated to give
4.6 g of crude product, which was recrystallized from 3:2
hexane:EtOAc to give 3.3 g (0.0131 mol, 68% based on the
7-benzyl-7-azabicyclo[2.2.1]heptane precursor13 of 7) of 8 as a
23 was produced by this microorganism just as it was
with A. ochraceous; the difference, therefore, does not lie
in the use of different microorganisms in the two experi-
ments. Clearly, there are subtle structural details of
enzymic active sites that we do not yet appreciate.
1
white solid: mp 120-121 °C; H NMR (CDCl3) δ 1.38 (dm, J
) 7.2 Hz, 4 H), 1.78 (dm, J ) 7.2 Hz, 4 H), 2.42 (s, 3 H), 4.18
(m, 2 H), 7.28 (d, J ) 8.0 Hz, 2 H), 7.80 (d, J ) 8.2 Hz, 2 H);
13C NMR (CDCl3) δ 21.4, 30.0, 59.1, 127.4, 129.3, 137.7, 143.2;
IR (mull) 604 (s), 674 (s), 1055 (s), 1088 (s), 1152 (s), 1199 (m),
1334 (s), 1608 (m), 2922 (s), 2956 (s) cm-1; GCMS m/ z 251
(M+). Anal. Calcd for C13H17NO2S: C, 62.12; H, 6.82; N, 5.58.
Found: C, 62.05; H, 6.82; N, 5.56.
Su m m a r y
The best chemical yield observed in this study of the
oxygenation of N-acyl-7-azabicyclo[2.2.1]heptanes (Table
1) is the oxygenation of 10 using R. nigricans to give a
combined 89% yield of 2-endo- and 2-exo-alcohols (15 and
2). Oxidation of both 15 and 2 will generate ketone 3 in
good yield. Since ketone 3 has been carried on to
epibatidine as described by Fletcher and co-workers,4 our
results provide a link between the synthesis of 7-benzyl-
7-azabicyclo[2.2.1]heptane described by Hassner and
Belostotskii13 and the conversion of ketone 3 into epiba-
tidine and thereby constitute a formal synthesis of the
unnatural enantiomer of this compound.
7-(P h en yloxyca r bon yl)-7-a za bicyclo[2.2.1]h ep ta n e (9).
A 250 mL separatory funnel was charged with crude salt 7
(6.20 g) and shaken vigorously with 2 N NaOH (100 mL).
Phenyl chloroformate (3.1 g, 20 mmol) was added, and the
funnel was vigorously shaken for 10-15 min. The aqueous
mixture was then extracted with CH2Cl2 (4 × 50 mL). The
organic phases were dried (Na2SO4) and concentrated to give
a solid crude product that was recrystallized from hexane to
give 2.5 g (0.0115 mol, 58% based on the precursor of 7) of 9
1
as a white solid: mp 70-71 °C; H NMR (CDCl3) δ 1.51 (d, J
) 6.8 Hz, 4 H), 1.90 (d, J ) 6.8 Hz, 4 H), 4.44 (s, 2 H), 7.13 (m,
2 H), 7.19 (m, 1 H), 7.36 (m, 2 H); 13C NMR (CDCl3) δ 30.2,
56.9, 122.0, 125.5, 129.6, 151.7, 153.8; IR (mull) 1042 (m), 1165
(s), 1196 (s), 1391 (s), 1718 (s), 2922 (s), 2956 (s) cm-1; GCMS
m/ z 217 (M+). Anal. Calcd for C13H15NO2: C, 71.87; H, 6.96,
6.45. Found: C, 71.77; H, 6.95; N, 6.40.
The best enantioselectivity observed is in the oxygen-
ation of 9 using B. bassiana to give the 2-endo-alcohol
14 with 51% ee and in 46% isolated yield. The predomi-
nant enantiomer of 14 is shown to have the 1S absolute
configuration, which is opposite to that required by
natural (-)-epibatidine. While these results demonstrate
the potential of microbiological oxygenations to deliver
enantioselective reactions, in this case they are neither
of the correct absolute configuration nor of high enough
enantioselectivity to be of use in the synthesis of either
natural (-)-epibatidine or its unnatural enantiomer.
In previous work with B. bassiana we have drawn
attention to a spatial arrangement in the products that
can be summarized as a trans orientation of the newly
introduced hydroxyl group(s) to an existing functional
group (usually an amide group) in the substrate.7 Such
an orientation is seen in the present results in the
hydroxylation of the 7-azabicyclo[2.2.1]heptane (products
14-16)18 and 8-azabicyclo[3.2.1]octane (23) systems. The
surprising result in this work is in the hydroxylation of
N-(phenyloxycarbonyl)-9-azabicyclo[3.3.1]nonane (22),
which produces the cis-3-alcohol 26.
Finally, oxygenation of the p-toluenesulfonamide 8 by
B. bassiana on the methyl group of the toluene rather
than on the bicyclic hydrocarbon ring system suggests
that the tosyl group finds easier entry into the active site
cavity of the hydroxylase than does the bicyclic system.
The bioconversion of other p-toluenesulfonamide deriva-
tives of saturated cyclic hydrocarbons using B. bassiana
has resulted in oxygenation of the saturated portion of
the substrate.19
7-[(1,1-Dim eth yleth oxy)ca r bon yl]-7-a za bicyclo[2.2.1]-
h ep ta n e (10). A 250 mL flask was charged with crude salt 7
(5.0 g), di-tert-butyl dicarbonate (7.0 g, 0.032 mol), dioxane (100
mL), and 2 N NaOH (50 mL) and stirred at rt for 24 h. The
solvents were removed by rotary evaporation. H2O (100 mL)
was added, and the product was extracted with CH2Cl2 (3 ×
75 mL). The CH2Cl2 phases were dried (Na2SO4) and concen-
trated to give an oily residue. Chromatography on SiO2 (5%
EtOAc:hexane, Rf 0.4) gave 2.2 g (0.0111 mol, 68%) of 10 as a
clear oil: 1H NMR (CDCl3) δ 1.37 (d, J ) 7.2 Hz, 4 H), 1.43 (s,
9 H), 1.74 (m, 4 H), 4.20 (bs, 2 H); 13C NMR (CDCl3) δ 28.66,
29.95, 56.50, 79.58. 156.17; IR (mull) 1090 (m), 1150 (s), 1185
(m), 1318 (m), 1369 (s), 1390 (m), 1702 (s), 2954 (m), 2975 (m)
cm-1; HRMS calcd 197.1416, obsd 197.1418. Anal. Calcd for
C
11H19NO2: C, 66.97; H, 9.71; N, 7.10. Found: C, 65.67; H,
9.61; N, 6.89.
7-(Ben zyloxycar bon yl)-7-azabicyclo[2.2.1]h eptan e (11).
A 250 mL separatory funnel containing crude salt 7 (5.0 g)
and 2 N NaOH (100 mL) was shaken for several minutes.
Benzyl chloroformate (3.4 g, 20.0 mmol) was added, and the
funnel was shaken vigorously for 10 min. The mixture was
extracted with CH2Cl2 (4 × 50 mL), and the combined CH2Cl2
phases were dried (Na2SO4) and concentrated to give a crude
oil. Chromatography on SiO2 (5% EtOAc-hexane, Rf 0.25)
gave 3.25 g (0.0140 mol, 85% based on the precursor to 7) of
11 as a colorless oil: 1H NMR (CDCl3) δ 1.42 (d, J ) 7.3 Hz,
4 H), 1.78 (m, 4 H), 4.31 (bs, 2 H), 5.12 (s, 2 H), 7.34 (m, 5 H);
13C NMR (CDCl3) δ 30.06, 56.57, 67.00, 128.17, 128.28, 128.8,
137.31, 155.96; IR 698 (m), 1093 (m), 1153 (s), 1260 (m), 1304
(m), 1313 (s), 1349 (m), 1394 (m), 1707 (s), 2953 (s) cm-1
HRMS calcd 231.1259, obsd 231.1261. Anal. Calcd for
14H17NO2: C, 72.70; H, 7.41; N, 6.06. Found: C, 72.09; H,
;
Exp er im en ta l Section
C
7.35; N, 6.06.
7-(p-Tolu en esu lfon yl)-7-a za bicyclo[2.2.1]h ep ta n e (8). A
250 mL separatory funnel was charged with crude salt 7 (6.20
g) and shaken vigorously with 100 mL of 2 N NaOH. p-
Toluenesulfonyl chloride (3.8 g, 20 mmol) was added, and the
funnel was vigorously shaken for 10-15 min. The aqueous
mixture was then extracted with CH2Cl2 (4 × 100 mL). The
7-Ben zoyl-7-a za bicyclo[2.2.1]h ep ta n e (12). A 250 mL
separatory funnel containing crude salt 7 (6.0 g) and 2 N
NaOH (100 mL) was shaken for several minutes. Phenyl
chloroformate (3.4 g, 20.0 mmol) was added, and the funnel
was shaken vigorously for 10 min. The mixture was extracted
with CH2Cl2 (4 × 50 mL), and the combined CH2Cl2 phases
were dried (Na2SO4) and concentrated to give a crude solid.
Chromatography on SiO2 (5% CH3CN-CH2Cl2, Rf 0.35) gave
2.32 g (0.0115 mol, 58% based on the precursor of 7) of 12 as
(18) Epimeric alcohols 15/2 and 16/17 may represent direct hy-
droxylation products or, alternately, could arise through a sequence
of enzymic conversions: (a) hydroxylation, (b) dehydrogenation, and
(c) reduction. We have no direct evidence to choose between the two
possibilities but note that the intermediate ketone of the latter
mechanism was not detected.
(19) (a) Fonken, G. S.; Herr, M. E.; Murray, H. C.; Reineke, L. M.
J . Org. Chem. 1968, 32, 3182. (b) J ohnson, R. A.; Herr, M. E.; Murray,
H. C.; Fonken, G. S. J . Org. Chem. 1968, 32, 3187.
1
a white solid: mp 75-76 °C; H NMR (CDCl3) δ 1.49 (d, J )
8.4 Hz, 4 H), 1.75-1.98 (m, 4 H), 4.12 (bs, 1 H), 4.75 (bs, 1 H),
7.42 (m, 3 H), 7.54 (m, 2 H); 13C NMR (CDCl3) δ 29.1, 30.9,
54.1, 59.1, 128.1, 128.6, 130.7, 136.7, 169.0; IR (mull) 1621
(s), 1615 (s), 1455 (s), 1445 (s), 1461 (m), 1576 (m), 2922 (s),

Oxygenation of Bridgehead Azabicycloalkanes
J . Org. Chem., Vol. 62, No. 7, 1997 2249
(s), 2952 (s), 3461 (s) cm-1
. Anal. Calcd for C13H15NO3: C,
2953 (s) cm-1. Anal. Calcd for C13H15NO: C, 77.58; H, 7.513;
N, 6.96. Found: C, 77.27; H, 7.38; N, 7.04.
66.94; H, 6.48; N, 6.01. Found: C, 67.02; H, 6.56; N, 5.97.
Alcohol 14 was converted to Mosher esters 18a and 18b;
Biotr a n sfor m a tion P r ocess. The cultures used in these
experiments were B. bassiana (ATCC 7159, UC-1365), R.
arrhizus (ATCC 11145, UC-4041), R. nigricans (UC-4285), and
A. ochraceus (NRRL 405, UC-4115). Fermentation media was
prepared by mixing cerelose (25 g/L) and Pharmamedia
(Archer Daniels Midland Co., 25 g/L) with tap water and
adjusting the pH to 7.2 with NH4OH. Fermentation tanks
containing 10 L of sterilized media (autoclaved at 180 °C for
90 min) were innoculated with 100 mL of the appropriate seed
culture (pregrown in a 500 mL flask for 3 days on two rotary
shakers at 250 rpm). Surfactant (SAG 471, Union Carbide)
(0.4 g/L) was added to the tank, the air flow adjusted to 12
L/min, and agitation was set at 250 rpm. The culture was
allowed to grow for 24 h at 28 °C, after which time the
substrate was added as a 0.1 g/mL acetone or DMF solution.
The fermentation or transformation was continued additional
days until the transformation was judged optimum by chro-
matographic analysis. The contents of the tank were trans-
ferred to a 25 L vat equipped with a drain and air-driven
propeller agitator. CH2Cl2 (8 L) was added, and the mixture
was agitated for 1-2 h. The contents were suction-filtered in
2 L portions through a layer of Celite on a polypropylene pad,
changing the Celite with each filtration. The CH2Cl2 was
separated, dried (Na2SO4), and concentrated by rotary evapo-
ration to give the crude residue, purified by column chroma-
tography as detailed below. Similarly, shake flask fermenta-
tions were carried out in 500 mL Erlenmeyer flasks equipped
with air-permeable microbe filters using 100 mL sterilized
media prepared as described above. The flasks were placed
on automated shakers (250 revolutions per min) at 28 °C for
the duration of the fermentations. No surfactant was added.
For extraction, 100 mL of CH2Cl2 was added to each flask, and
the flasks were shaken for 1 h. To remove the biomass,
collective filtration of the flask contents through Celite was
carried out. The organic phase was separated, dried, and
concentrated. Fermentations were monitored by removal and
workup of a sample aliquot (ca. 25 mL) followed by GCMS and
TLC analysis of the crude residue.
P r ep a r a t ion of 7-[(p -H yd r oxym et h yl)b en zen esu lfo-
n yl]-7-a za bicyclo[2.2.1]h ep ta n e (13) by Biocon ver sion of
7-(p-Tolu en esu lfon yl)-7-a za bicyclo[2.2.1]h ep ta n e (8) Us-
in g B. ba ssia n a . Bioconversion of 8 (1.00 g, 3.98 mmol) using
B. bassiana in a 10 L tank was carried out for 5 days.
Following workup (see general procedure above), the crude
residue was twice chromatographed on SiO2 [(1) 5% CH3CN-
CH2Cl2, Rf 0.25; (2) 40% acetone-hexane, Rf 0.40] to give 0.572
g (0.00205 mol, 54%) of 13 as a white solid: mp 121-122 °C;
1H NMR (CDCl3) δ 1.37 (d, J ) 7.2 Hz, 4 H), 1.75 (d, J ) 7.2
Hz, 4 H), 2.58 (s, 1 H), 4.16 (s, 2 H), 4.75 (s, 2 H), 7.44 (d, J )
8.3 Hz, 2 H), 7.82 (d, J ) 8.3 Hz, 2 H); 13C NMR (CDCl3) δ
30.5, 59.7, 64.5, 127.2, 128.0, 139.8, 146.6; IR (mull) 677 (s),
1054 (s), 1091 (s), 1150 (s), 1320 (s), 1412 (s), 1458 (s), 2925
(s), 2955 (s), 3493 (s) cm-1; GCMS m/ z 267 (M+), 238, 171,
107, 77, 68, 41. Anal. Calcd for C13H17NO3S: C, 58.41; H,
6.41; N, 5.24. Found: C, 58.55; H, 6.47; N, 5.22.
1
the esters were analyzed by 19F NMR (51.9% ee) and H NMR
(50.0% ee) spectroscopy.
P r ep a r a t ion of Mosh er E st er s 18a a n d 18b . Mosher
esters 18a and 18b were prepared by the general procedure
described below for 20a and 20b. The mixture of diastereoi-
somers was purified by chromatography over SiO2 (10%
EtOAc-hexane, Rf
0.15) to give a viscous oil: 1H NMR (CDCl3)
δ 18a 3.54 (OCH3), 18b 3.58 (OCH3); 19F NMR (CDCl3 vs
internal CFCl3 standard) 18a -71.93 (s, CF3), 18b -71.85 (s,
CF3).
P r ep a r a t ion of [(1,1-Dim et h ylet h oxy)ca r b on yl]-7-
a za bicyclo[2.2.1]h ep ta n -2-en d o-ol (15) by Biocon ver sion
of [(1,1-Dim et h ylet h oxy)ca r b on yl]-7-a za b icyclo[2.2.1]-
h ep ta n e (10) Usin g B. ba ssia n a . Bioconversion of 10 (1.00
g, 5.06 mmol) using B. bassiana in a 10 L tank was carried
out for 4 days. Following workup (see general procedure
above), the crude residue was twice chromatographed on SiO2
[(1) 5% MeCN-CH2Cl2, Rf 0.20; (2) 10% acetone-hexane, Rf
0.1] to give 0.111 g (0.00052 mol, 10%) of 15 as a viscous oil:
1H NMR (CDCl3) δ 1.05 (dd, J ) 12.7, 3.5 Hz, 1 H), 1.41 (s, 9
H), 1.45-1.75 (m, 3 H), 2.65 (bs, 1 H), 4.09 (m, 2 H), 4.30 (m,
1 H); 13C NMR (CDCl3) δ 20.66, 28.26, 29.79, 39.07, 57.38,
60.00, 70.64, 79.73, 155.69; GCMS m/ z 213 (M+), 157, 113,
69, 57, 41; [R]25 +1.2° (c ) 1.00, CHCl3).
D
Alcohol 15 was converted to Mosher esters 19a and 19b
followed by 19F NMR (25.2% ee) and 1H NMR (26.7% ee)
analysis. Swern oxidation of a sample of 15 (0.082 g) gave
ketone 3, [R]D +19.0° (c ) 1.00, CHCl3).
P r ep a r a t ion of Mosh er E st er s 19a a n d 19b . Mosher
esters 19a and 19b were prepared by the general procedure
described below for 20a and 20b. The mixture of diastereoi-
somers was purified by chromatography over SiO2 (20%
EtOAc-hexane, Rf 0.45) to give a viscous oil: 1H NMR (CDCl3)
δ 19a 3.5 (OCH3), 19b 3.55 (OCH3); 19F NMR (CDCl3 vs
internal CFCl3 standard) 19a -71.99 (s, CF3), 19b -71.90 (s,
CF3).
P r ep a r a tion of 15 a n d [(1,1-Dim eth yleth oxy)ca r bo-
n yl]-7-a za bicyclo[2.2.1]h ep ta n -2-exo-ol (2) by Biocon ver -
sion of [(1,1-Dim et h ylet h oxy)ca r b on yl]-7-a za b icyclo-
[2.2.1]h ep ta n e (10) Usin g R. n igr ica n s. Bioconversion of
10 (0.250 g, 1.27 mmol) using R. nigricans in 10 shake flasks
was carried out for 3.5 days. Following workup, the residue
was first chromatographed on SiO2 (10% acetone-CH2Cl2, Rf
0.30) to give a mixture of endo and exo isomers that were then
separated on SiO2 (40% EtOAc-hexane, Rf endo 0.20, Rf exo
0.10) to give 0.169 g (0.793 mmol, 62%) of endo isomer 15 and
0.075 g (0.352 mmol, 27%) of exo isomer 2, both as viscous
oils:
1H and 13C NMR (CDCl3) for 15 are as given above. Exo
isomer 2: 1H NMR (CDCl3) δ 1.20-1.29 (m, 2 H), 1.44 (s, 9
H), 1.57-1.73 (m, 3 H), 1.82 (dd, J ) 13.1, 6.8 Hz, 1 H), 1.95
(bs, 1 H), 3.85 (dm, J ) 5.7 Hz, 1 H), 4.10 (d, J ) 5.1 Hz, 1 H),
4.22 (t, J ) 4.6 Hz, 1 H); 13C NMR (CDCl3) δ 23.97, 28.15,
28.30, 42.52, 55.42, 63.19, 74.41, 79.90, 156.68; [R]25D +8.8° (c
) 0.639, CH2Cl2); GCMS m/ z 213 (M+), 157, 113, 69, 68, 57,
41.
P r ep a r a t ion of 7-(P h en yloxyca r bon yl)-7-a za bicyclo-
[2.2.1]h ep ta n -2-en d o-ol (14) by Biocon ver sion of 7-(P h e-
n yloxyca r bon yl)-7-a za bicylo[2.2.1]h ep ta n e (9) Usin g B.
ba ssia n a . Bioconversion of 9 (1.00 g, 4.60 mmol) using B.
bassiana in a 10 L tank was carried out for 5 days. Following
workup (see general procedure above), the crude residue was
chromatographed three times on SiO2 [(1) 20% acetone-
hexane, Rf 0.20; (2) 95:4.5:0.5 CHCl3:MeOH:Et3N, Rf 0.25; (3)
30% acetone-hexane, Rf 0.40] to give 0.495 g (0.00212 mol,
46%) of 14 as a white solid: mp 103-104 °C; 1H NMR (50 °C,
CDCl3) δ 1.13 (dd, J ) 12.7, 3.4 Hz, 1 H), 1.61 (ddd, J ) 11.4,
9.0, 3.9 Hz, 1 H), 1.75 (dddd, J ) 12.3, 3.6, 3.1, 1.5 Hz, 1 H),
1.89 (m, 1 H), 1.9 (s, 1 H), 2.25 (m, 1 H), 2.30 (m, 1 H), 4.31
(dd, J ) 4.5, 3.8 Hz, 1 H), 4.33 (dd, J ) 4.5, 3.8 Hz, 1 H), 4.40
(dddd, J ) 9.7, 4.8, 3.4, 1.5 Hz, 1 H), 7.08 (m, 2H), 7.16 (m,
1H), 7.36 (m, 2H); 13C NMR (50 °C) (CDCl3) δ 20.8, 29.9, 39.3,
57.7, 60.2, 70.6, 121.4, 125.1, 129.1, 152.2, 153.4; [R]25D + 5.2°
(c ) 1.10, CHCl3); GCMS m/ z 234 (M+ + H), 140, 94, 79, 41;
IR (mull) 1042 (s), 1201 (s), 1380 (s), 1706 (s), 1720 (s), 2924
Alcohol 15 was converted to Mosher esters 19a and 19b;
19F NMR analysis: 29.3% ee. Swern oxidation of 15 (0.050 g)
gave ketone 3, [R]D +20.7° (c ) 1.00, CHCl3).
P r ep a r a tion of 7-(Ben zyloxyca r bon yl)-7-a za bicyclo-
[2.2.1]h ep ta n -2-en d o-ol (16) by Biocon ver sion of 7-(Ben -
zyloxyca r bon yl)-7-a za bicyclo[2.2.1]h ep ta n e (11) Usin g B.
ba ssia n a . Bioconversion of 11 (1.30 g, 5.62 mmol) using B.
bassiana in a 10 L tank was carried out for 6.5 days. Following
workup (see general procedure above), the residue was chro-
matographed three times on SiO2 [(1) 20% EtOAc-hexane, Rf
0.30; (2) 15% acetone-hexane, Rf 0.10; (3) 12% acetone-CH2-
Cl2, Rf 0.30) to give 0.388 g (0.00157 mol, 28%) of 16 as a
viscous oil: 1H NMR (CDCl3) δ 1.07 (dd, J ) 12.7, 3.4 Hz, 1
H), 1.49-1.66 (m, 2 H), 1.77 (m, 1 H), 2.14-2.22 (m, 2 H), 2.50
(m, 1 H), 4.22 (m, 2 H, bridgehead H), 4.30 (m, 1 H, carbinol
H), 5.09 (s, 2 H, benzylic H), 7.33 (s, 5 H, aromatic); 13C NMR
(CDCl3) δ 20.78, 29.91, 39.18, 57.40, 59.99, 66.90, 70.66, 127.84,
128.07, 128.52, 136.64, 155.57; IR (mull) 3434 (m), 2908 (m),

2250 J . Org. Chem., Vol. 62, No. 7, 1997
Davis et al.
1706 (s), 1682 (s), 1455 (m), 1406 (s), 1352 (s), 1302 (s), 1099
(0.295 g, 1.4 mmol) was added, and the mixture was stirred 6
h at rt. The solvents were evaporated, and the residue was
diluted with 15 mL of CH2Cl2 and washed with 10 mL of H2O.
The aqueous phase was back-extracted with CH2Cl2 (10 mL),
and the combined organic phases were dried (Na2SO4) and
concentrated. The residue was chromatographed on SiO2 (10%
acetone-hexane, Rf 0.15) to give 0.164 g (0.77 mmol, 85%) of
15 as a viscous oil (1H, 13C NMR purity >95%).
(s), 1086 (s) cm-1; [R]25 -0.3° (c ) 1.00, CHCl3); HRMS calcd
D
(M+ + H) 248.1287, obsd 248.1289. Anal. Calcd for C14H17
-
NO3: C, 68.00; H, 6.93; N, 5.66. Found: C, 66.79; H, 6.83; N,
5.45.
Alcohol 16 was converted to Mosher esters 20a and 20b as
described below, and the esters were submitted to 19F (4.1%
ee) and 1H (5.9% ee) NMR analysis.
Swern oxidation of a sample of 15 (0.162 g, derived from
14) gave ketone 3, [R]D +37.7° (c ) 1.00, CHCl3).
Gen er a l P r oced u r e for P r ep a r a tion of Mosh er Ester s.
P r ep a r a tion of 20a ,b fr om 16. To a 1 mL reaction vial
equipped with a septum and spin vane were added CHCl3 (0.5
mL), Et3N (13 µL,0.1 mmol), 16 (0.008 g, 0.032 mmol), and a
single crystal of DMAP. (R)-(-)-R-Methoxy-R-(trifluorometh-
yl)phenylacetyl chloride (0.0164 g, 0.064 mmol) was then added
via µL syringe (either neat or as a CDCl3 solution), and the
mixture was stirred overnight at rt. The contents of the
reaction were eluted through a short column of SiO2 (20%
EtOAc-hexane) to give a purified mixture of diastereomers
20a and 20b as a viscous oil: 1H NMR (CDCl3) δ 20a 3.51
(OCH3), 20b 3.54 (OCH3); 19F NMR (CDCl3 vs internal CFCl3
standard) 20a -71.96 (s, CF3), 20b -71.89 (s, CF3).
Con ver sion of 7-(Ben zyloxyca r b on yl)-7-a za b icyclo-
[2.2.1]h ep ta n -en d o-2-ol (16) to 7-[(1,1-Dim eth yleth oxy)-
ca r bon yl]-7-a za bicyclo[2.2.1]h ep ta n -en d o-2-ol (15). A 15
mL flask equipped with a magnetic stir bar was charged with
10% Pd-C (0.10 g) under a stream of N2 followed by addition
of MeOH (5 mL), 16 (from bioconversion of 11 using B.
bassiana, 0.050 g, 0.202 mmol), and ammonium formate (0.063
g, 1.0 mmol). The mixture was stirred for 3 h at rt and then
filtered through a layer of Celite. The MeOH was removed
by evaporation, and the residue was diluted with dioxane (2
mL) and H2O (1 mL) followed by addition of NaOH (0.10 g)
and (Boc)2O (0.088 g, 0.4 mmol). The mixture was vigorously
stirred overnight at rt. The solvents were removed by evapo-
ration, and the residue was diluted with 10 mL of CH2Cl2 and
washed with an equal volume of H2O. The aqueous phase was
back-extracted with 10 mL of CH2Cl2, and the combined CH2-
Cl2 phases were dried (Na2SO4) and concentrated. The residue
was chromatographed on SiO2 (10% acetone-hexane, Rf 0.15)
to give 0.021 g (0.098 mmol, 48%) of 15: [R]25D -0.3° (c ) 1.00,
CHCl3). Alcohol 15 was converted to Mosher esters 19a and
P r ep a r a tion of 7-(Ben zyloxyca r bon yl)-7-a za bicyclo-
[2.2.1]h ep ta n -en d o-2-ol (16) a n d 7-(Ben zyloxyca r bon yl)-
7-a za bicyclo[2.2.1]h ep ta n -2-(exo)-ol (17) by Biocon ver -
sion of 7-(Ben zyloxycar bon yl)-7-azabicyclo[2.2.1]h eptan e
(11) Usin g R. n igr ica n s. Bioconversion of 11 (0.125 g, 0.540
mmol) using R. nigricans in 10 shake flasks was carried out
for 4.5 days. Following workup (see general procedure above)
the crude extract residue was chromatographed on SiO2 (40%
EtOAc-hexane, Rf endo-16 0.30, Rf exo-17 0.20). The endo
isomer 16 was purified further by chromatography on SiO2
(10% MeCN-CH2Cl2, Rf 0.25) to give 0.020 g (0.81 mmol,
15%): 1H, 13C NMR as given previously; [R]25D +1.5° (c ) 1.00,
CHCl3). Endo alcohol 16 was converted to Mosher esters 20a
1
19b as described above followed by 19F (6.7% ee) and H (5.5%
ee) NMR analysis.
Con ver sion of 7-(Ben zyloxyca r b on yl)-7-a za b icyclo-
[2.2.1]h ep ta n -exo-2-ol (17) to 7-[1,1-Dim eth yleth oxy)ca r -
bon yl]-7-a za bicyclo[2.2.1]h ep ta n -exo-2-ol (2). Following
the procedure described for conversion of 16 to 15, 17 (from
bioconversion of 11 using R. arrhizus, 0.030 g, 0.12 mmol),
10% Pd-C (0.030 g), MeOH (5 mL), ammonium formate (0.038
g, 0.60 mmol), NaOH (0.08 g), and (Boc)2O (0.050 g) were used
for conversion to 2. After workup, chromatography on SiO2
(40% EtOAc-hexane, Rf 0.25) gave 0.016 g (0.075 mmol, 63%)
1
and 20b as described above followed by 19F (36.5% ee) and H
(35.1% ee) NMR analysis.
The exo isomer 17 was further chromatographed on SiO2
(10% MeCN-CH2Cl2, Rf 0.20) to give 0.012 g (0.48 mmol, 8%)
as a viscous oil: 1H NMR (CDCl3) δ 1.22-1.30 (m, 2 H), 1.58-
1-77 (m, 3 H), 1.83 (dd, J ) 13.1, 6.8 Hz, 1 H), 2.2 (b, 1 H),
3.88 (d, J ) 5.9 Hz, 1 H), 4.20 (d, J ) 5.2 Hz, 1 H), 4.34 (t, J
) 4.5 Hz, 1 H), 5.11 (s, 2 H), 7.25-7.35 (m, 5 H); 13C NMR
(CDCl3) δ 24.03, 28.33, 42.26, 55.37, 63.22, 66.96, 74.21, 127.91,
128.04, 128.50, 136.66, 156.52; HRMS calcd for C14H17NO3 (M+
+ H) 248.1287, obsd 248.1289; [R]25D +11.6° (c ) 1.00, CHCl3).
P r ep a r a tion of 7-(Ben zyloxyca r bon yl)-7-a za bicyclo-
[2.2.1]h ep ta n -en d o-2-ol (16) a n d 7-(Ben zyloxyca r bon yl)-
7-a za bicyclo[2.2.1]h ep ta n -2-exo-ol (17) by Biocon ver sion
of 7-(Ben zyloxyca r bon yl)-7-a za bicyclo[2.2.1]h ep ta n e (11)
u sin g R. a r r h izu s. Bioconversion of 11 (0.500 g, 2.16 mmol)
using R. arrhizus was carried out in 32 shake flasks for 3.5
days. Two isomeric products were separated by chromatog-
raphy on SiO2 (15% MeCN-CH2Cl2, Rf endo-16 0.20, Rf exo-
17 0.15). Isomer 16 was further chromatographed on SiO2
(50% EtOAc-hexane, Rf 0.30) to give 0.028 g (5%): [R]25D +1.0°
(c ) 1.00, CHCl3). The endo isomer 16 was converted to
of 2: [R]25 +4.5° (c ) 0.639, CH2Cl2).
D
Con ver sion of 7-[(1,1-Dim et h ylet h oxy)ca r b on yl]-7-
a za bicyclo[2.2.1]h ep ta n -en d o-2-ol (15) to 7-[(1,1-Dim eth -
yleth oxy)ca r bon yl]-7-a za bicyclo[2.2.1]h ep ta n -2-on e (3).
To a 25 mL two-neck flask equipped with a magnetic stir bar,
N2
source, and low-temperature thermometer was added CH2-
Cl2 (7 mL) followed by oxalyl chloride (0.045 g, 0.35 mmol).
The solution was cooled to -78 °C via dry ice/2-propanol bath.
DMSO (0.055 g in 1 mL of CH2Cl2, 0.70 mmol) was added via
syringe, maintaining the temperature at <-60 °C, and the
solution was stirred for an additional 10 min at -70 °C. A 1
mL CH2Cl2 solution of 15 (0.050 g, 0.23 mmol) was added
dropwise at -70 °C, after which the solution was stirred for
an additional 20 min. The reaction was quenched with Et3N
(0.12 g, 3.9 mmol), and the mixture was allowed to warm to
rt. The mixture was washed with H2O (10 mL), and the
aqueous phase was back-extracted with an equal volume of
CH2Cl2. The combined CH2Cl2 phases were dried (Na2SO4)
and concentrated. Chromatography on SiO2 (5% acetone-
hexane, Rf 0.15) gave 0.038 g (0.18 mmol, 77%) of 3 as a clear
oil: 1H NMR (CDCl3) δ 1.44 (s, 9 H), 1.51-1.67 (m, 2 H), 1.96-
2.02 (m, 3 H), 2.46 (dd, J ) 17.4, 5.3 Hz, 1 H), 4.23 (d, J ) 4.8
Hz, 1 H), 5.46 (1, J ) 4.5 Hz, 1 H); 13C NMR (CDCl3) δ 24.43,
27.56, 28.20, 45.24, 56.05, 63.94, 80.84, 140.34, 155.06; GCMS
m/ z 211 (M+), 127, 83, 68, 57, 41.
P r ep a r a t ion of 8-(P h en yloxyca r bon yl)-8-a za b icyclo-
[3.2.1]octa n e (21). A 25 mL flask equipped with a septum
was charged with tropane (2.5 g, 20 mmol) and CH2Cl2 (15
mL). The solution was cooled to 5 °C with an ice bath, and a
5 mL CH2Cl2 solution of phenyl chloroformate (3.10 g, 19.8
mmol) was added dropwise via syringe over 5 min. The ice
bath was removed, and the solution was stirred overnight at
rt. The solution was washed with 0.1 N NaOH (15 mL) and 1
N HCl (15 mL), and the aqueous phases were each back-
extracted with CH2Cl2 (10 mL). The combined CH2Cl2 phases
Mosher esters 20a and 20b as described above followed by 19
F
(32.4% ee) and 1H (30.2% ee) NMR analysis.
The isomer 17 was chromatographed twice on SiO2 [(1) 15%
acetone-CH2Cl2, Rf 0.30; (2) 50% EtOAc-hexane, Rf 0.20] to
give 0.098 g (0.39 mmol, 18%) of pure material; [R]25D +4.4° (c
) 1.00, CHCl3).
Con ver sion of 7-(P h en yloxyca r bon yl)-7-a za bicyclo-
[2.2.1]h ep ta n -en d o-2-ol (14) to 7-[(1,1-Dim eth yleth oxy)-
ca r bon yl]-7-a za bicyclo[2.2.1]h ep ta n -en d o-2-ol (15). A 15
mL flask equipped with a magnetic stir bar and cold water
condenser was charged with 14 (from bioconversion of 9 using
B. bassiana, 0.210 g, 0.901 mmol), EtOH (3 mL), H2O (1.2 mL),
NaOH (0.2 g, 5 mmol), and 18-crown-6 ether (0.026 g, 0.1
mmol) and refluxed for 18 h. The reaction was monitored by
TLC for consumption of starting material (30% acetone-
hexane, Rf 0.4), and when complete, the pH was carefully
adjusted to 1-2 with concd HCl and the mixture refluxed for
an additional 1 h and cooled to rt. The solvents were removed
by rotary evaporation, and the residue was diluted with
dioxane (4 mL), H2O (2 mL), and NaOH to pH 10-11. (Boc)2O

Oxygenation of Bridgehead Azabicycloalkanes
J . Org. Chem., Vol. 62, No. 7, 1997 2251
were dried (Na2SO4) and concentrated to give a crude solid
residue that was recrystallized from hexane to give 2.77 g (12
mmol, 60%) of 21 as a white solid: mp 70-71 °C; 1H NMR
(CDCl3) δ 1.4-2.1 (m, 10 H), 4.34 (m, 1 H), 4.42 (m, 1 H), 7.15
(m, 3 H), 7.35 (m, 2 H); 13C NMR (CDCl3) δ 16.7, 27.6, 28.3,
30.6, 31.6, 54.2, 54.8, 121.7, 125.0, 129.2, 151.4, 151.6; GCMS
m/ z 231 (M+), 158, 138, 110, 95, 77, 67, 55; IR (mull) 2926 (s),
MeCN-CH2Cl2, Rf 0.20) to give 0.52 g (2.1 mmol, 60%) of 23
1
as a white solid: mp 153-154 °C (lit.16 mp 151-153 °C); H,
13C NMR data are in agreement with that given for 23 above.
P r ep a r a t ion of 9-(P h en yloxyca r bon yl)-9-a za b icyclo-
[3.3.1]n on a n -3-on e (27) a n d 9-(P h en yloxyca r bon yl)-9-
a za bicyclo[3.3.1]n on a n -exo-3-ol (26) by Biocon ver sion of
9-(P h en yloxycar bon yl)-9-azabicyclo[3.3.1]n on an e (22) Us-
in g B. ba ssia n a . Bioconversion of 22 (0.90 g, 3.7 mmol) using
B. bassiana was carried out in a 10 L tank for 3.5 days.
Following workup (see general procedure above), the crude
residue was chromatographed on SiO2 (5% MeCN-CH2Cl2, Rf
27 ) 0.35). After elution of the ketone 27, the alcohol was
eluted with EtOAc (Rf 26 ) 0.10). The ketone product was
then twice chromatographed on SiO2 [(1) 20% EtOAc-hexane,
Rf ) 0.15; (2) 5% MeCN-CH2Cl2, Rf ) 0.35] to give 0.147 g
(0.567 mmol, 15%) of 27 as a pale yellow solid: mp 125 °C; 1H
NMR (CDCl3) δ 1.61-1.70 (m, 4 H), 1.87-1.93 (m, 2 H), 2.46
(dd, J ) 16.3, 8.2 Hz, 2 H, -COCH2-), 2.75 (dd, J ) 16.6, 7.0
Hz, 2 H, -COCH2-), 4.83 (bs, 1 H, bridgehead H), 4.92 (bs, 1
H, bridgehead H), 7.14 (m, 2 H), 7.24 (m, 1 H), 7.38 (m, 2 H);
13C NMR (CDCl3) δ 16.18, 30.72, 30.79, 45.61, 45.24, 48.16,
48.95, 121.64, 125.33, 129.39, 151.20, 152.84, 208.55; IR (mull)
2954 (s), 1717 (s), 1702 (s), 1411 (s), 1367 (s), 1346 (s), 1341
(s), 1207 (s), 1048 (s) cm-1; HRMS calcd 259.1208, obsd
259.1209. Anal. Calcd for C15H17NO3: C, 69.48; H, 6.61; N,
5.40. Found: C, 69.42; H, 6.50; N, 5.45.
1710 (s), 1410 (s), 1206 (s), 1175 (m), 1034 (m), 1325 (m) cm-1
.
Anal. Calcd for C14H17NO2: C, 72.70; H, 7.41; N, 6.06.
Found: C, 72.76; H, 7.14; N, 5.98.
P r ep a r a t ion of 9-(P h en yloxyca r b on yl)-9-a za bicyclo-
[3.3.1]n on a n e (22). 9-Methyl-9-azabicyclo[3.3.1]nonane was
prepared by Wolff-Kishner reduction of pseudopelletierine as
described in the literature.15 Demethylation was carried out
as described above for the preparation of 21 and as reported
in the literature, using 9-methyl-9-azabicyclo[3.3.1]nonane
(2.20 g, 15.8 mmol), CH2Cl2 (20 mL), and a 10 mL CH2Cl2
solution of phenyl chloroformate (3.1 g, 20 mmol). Following
workup, the crude residue was recrystallized from hexane to
yield 1.1 g (4.5 mmol, 28%) of 22 as a white solid: mp 99-100
1
°C (lit.15 mp 100-102 °C); H NMR (CDCl3) δ 1.75 (m, 6 H),
1.90-2.21 (m, 6 H), 4.38 (s, 1 H), 4.47 (s, 1 H), 7.14 (m, 3 H),
7.36 (m, 2 H); 13C NMR (CDCl3) δ 20.5, 29.7, 30.3, 46.8, 47.9,
122.1, 125.3, 129.5, 152.1, 153.3.
P r ep a r a t ion of 8-(P h en yloxyca r bon yl)-8-a za b icyclo-
[3.2.1]octa n -en d o-3-ol (23) a n d 8-(P h en yloxyca r bon yl)-
8-a za bicyclo[3.2.1]octa n -3-on e (24) by Biocon ver sion of
8-(P h en yloxyca r bon yl)-8-a za bicyclo[3.2.1]octa n e (21) Us-
in g A. och r a ceu s. Bioconversion of 21 (0.800 g, 3.46 mmol)
using A. ochraceous was carried out in 80 shake flasks for 3
days. Following workup (see general procedure above), the
residue was chromatographed on SiO2 (5% MeCN-CH2Cl2, Rf
alcohol 23 0.10, Rf ketone 24 0.45) to give 0.31 g (1.25 mmol,
The alcohol product was then twice chromatographed on
SiO2 [(1) 20% MeCN-CH2Cl2, Rf ) 0.30; (2) 30% acetone-
hexane, Rf ) 0.40] to give 0.290 g (1.11 mmol, 30%) of 26 as a
white solid: mp 116-118 °C; 1H NMR (CDCl3) δ 1.60-1.85
(m, 7 H), 1.80-1.98 (m, 2 H), 2.04-2.15 (6 line m, 2 H), 4.50
(m, 1 H), 4.63 (m, 2 H), 7.10 (m, 2 H), 7.16 (m, 1 H), 7.35 (m,
2 H); 13C NMR (CDCl3) δ 20.76, 29.22, 29.81, 39.98, 40.43,
47.71, 48.58, 65.22, 122.09, 125.53, 129.61, 151.83, 152.95;
HRMS calcd 261.1365, found 261.1371; IR (mull) 3454 (m),
2954 (s), 1693 (s), 1424 (s), 1203 (s), 1349 (w), 1114 (m), 1061
36%) of 23 as a pale yellow solid (1H NMR purity 95%).
A
portion of alcohol 23 (0.150 g) was further chromatographed
on SiO2 (90:9:1 EtOAc:MeOH:Et3N, Rf 0.65) to give 0.136 g of
23: mp 151-153 °C (lit.16 mp 151-153 °C); 1H NMR (CDCl3)
δ 1.72-1.82 (m, 2 H), 1.91-2.27 (m, 6 H), 4.00 (t, J ) 4.9 Hz,
3 H), 4.33 (bs, 1 H), 4.42 (bs, 1 H), 7.15-7.39 (m, 1 H); 13C
NMR (CDCl3) δ 27.72, 28.45, 38.16, 39.08, 52.91, 53.49, 64.85,
121.59, 125.03, 129.13, 151.19, 151.53; IR (mull) 3475 (s), 2924
(s), 1694 (s), 1426 (s), 1361 (m), 1206 (s), 1194 (s), 1043 (s)
cm-1. Anal. Calcd for C14H17NO3: C, 68.00; H, 6.93; N, 5.66.
Found: C, 67.95; H, 6.90; N, 5.72.
(s) cm-1
. Anal. Calcd for C15H19NO3: C, 68.94; H, 7.33; N,
5.36. Found: C, 68.88; H, 7.43; N, 5.36.
Con ver sion of 9-(P h en yloxyca r bon yl)-9-a za bicyclo-
[3.3.1]n on a n -exo-3-ol (26) to 9-Ben zoyl-9-a za bicyclo[3.3.1]-
n on a n -exo-3-ol (28). A 10 mL flask equipped with a stir bar
and cold H2O condenser was charged with 26 (0.090 g, 0.344
mmol), EtOH (4 mL), H2O (1.5 mL), 18-crown-6 ether (0.015
g, 0.057 mmol), and NaOH (0.3 g, 7.5 mmol). The solution
was refluxed for 2 h, after which time consumption of 26 was
noted by TLC (Rf 26 ) 0.35, 50% EtOAc-hexane). After the
solution was cooled to 50 °C, the pH was adjusted to 1-2 with
concd HCl and refluxing was continued for an additional 30
min. The reaction mixture was cooled to rt, and the solvents
were removed by rotary evaporation. The residue was diluted
with H2O (4 mL), and the pH was adjusted to 10-11 with
NaOH, followed by addition of benzoyl chloride (0.141 g, 1.0
mmol). The mixture was stirred vigorously for 2 h at rt,
diluted with H2O (5 mL), and extracted with CH2Cl2 (3 × 10
mL). The combined CH2Cl2 phases were dried (Na2SO4) and
concentrated to give a crude viscous residue. The residue was
chromatographed on SiO2 (75% EtOAc-hexane, Rf ) 0.15) to
give 0.023 g (0.093 mmol, 27%) of 28 as a white solid, mp 145-
149 °C. Recrystallization from acetone-hexane gave crys-
tals: mp 153-154 °C (lit.17 mp 152-153 °C); 1H NMR (CDCl3)
δ 1.50-2.15 (4 multiplets, 11 H), 3.95 (bs, 1 H), 4.63 (7 line
m, 1 H), 4.97 (bs, 1 H), 7.39 (s, 5 H); 13C NMR (CDCl3) δ 20.53,
28.94, 30.06, 39.66, 40.45, 44.73, 51.06, 64.93, 126.30, 128.57,
129.53, 136.32, 169.24; GCMS m/ z 245 (M+), 228, 200, 140,
122, 105, 96, 77, 55, 27.
The ketone 24 was then twice chromatographed on SiO2 [(1)
15% EtOAc-hexane, Rf 0.10; (2) 5% MeCN-CH2Cl2, Rf 0.45]
to give 0.031 g (0.12 mmol, 4%) of a white solid: mp 122-123
°C; 1H NMR (CDCl3) δ 1.78 (m, 2 H), 2.21 (m, 2 H), 2.47 (m, 2
H), 2.79 (m, 2 H), 4.74 (m, 2 H), 7.15 (dd, J ) 5.4, 1.1 Hz, 2
H), 7.21 (m, 1 H), 7.39 (t, J ) 7.4 Hz, 2 H); 13C NMR (CDCl3)
δ 28.66, 29.47, 48.63, 49.38, 53.44, 121.56, 125.56, 129.39,
150.96, 151.83, 207.45; IR (mull) 2924 (s), 1721 (s), 1707 (s),
1402 (s), 1342 (s), 1321 (m), 1282 (m), 1058 (s), 1000 (m) cm-1
;
HRMS found 245.1063, C14H15NO3 requires 245.1052, 152, 109,
94, 81, 67 m/ z. Anal. Calcd for C14H15NO3: C, 68.56; H, 6.17;
H, 5.71. Found: C, 68.54; H, 6.24; 5.71.
Con ver sion of Tr op in e (25) to 8-(P h en yloxyca r bon yl)-
8-a za bicyclo[3.2.1]octa n -en d o-3-ol (23). Preparation of 23
was carried out according to the literature procedure17 with
the following modifications. A 25 mL flask was charged with
tropine (25, 0.50 g, 3.5 mmol) and CH2Cl2 (15 mL). Phenyl
chloroformate (1.2 g, 7.6 mmol) was added, and the solution
was stirred for 24 h at rt. The reaction mixture was evapo-
rated to dryness, diluted with 2 N HCl (20 mL), and extracted
with Et2O (3 × 75 mL). The combined Et2O phases were
washed with saturated NaHCO3 (25 mL), dried (Na2SO4), and
concentrated. The residue was chromatographed on SiO2 (10%
J O962192F