Nucleophilic Alkylation on Anti-Bredt Iminium Ions. Facile Entry to the Synthesis of 1-Alkylated 2-Azabicyclo[3.3.1]nonanes (Morphans) and 5-Azatricyclo[6.3.1.01,5]dodecane

Full text of DOI:10.1021/jo9715579

8280
J . Org. Chem. 1997, 62, 8280-8281
Sch em e 1
Nu cleop h ilic Alk yla tion on An ti-Br ed t
Im in iu m Ion s. F a cile En tr y to th e
Syn th esis of 1-Alk yla ted
2-Aza bicyclo[3.3.1]n on a n es (Mor p h a n s) a n d
5-Aza tr icyclo[6.3.1.01,5]d od eca n e
Naoki Yamazaki, Hidetaka Suzuki, and
Chihiro Kibayashi*
organometallic reagents on hydrazine N,O-acetals. From
these results it was envisioned, as outlined in eq 1, that
School of Pharmacy, Tokyo University of Pharmacy & Life
Science, Horinouchi, Hachioji, Tokyo 192-0392, J apan
Received August 21, 1997
The 2-azabicyclo[3.3.1]nonane (morphan) ring system
2 (R ) H) is present in morphine (1) and related synthetic
compounds with analgesic activity such as levorphanol
and pentazocine and also found in a great number of
indole alkaloids.1 Very recently, the discovery and
structural determination of the novel immunosupressant
FR901483 (3), isolated from the fermentation broth of
Cladobotryum sp. No. 11231, was reported by the Fujisa-
wa group.2 From a structural point of view, the most
using appropriate tricyclic N,O-acetals 5 such alkylation
protocol could provide a new method for the alkylation
of the bridgehead position proceeding by a pathway
involving an SN1 reaction on intermediary bridgehead
iminium ions 6 with the carbon placed at the bridgehead.
Despite Bredt’s rule,5 a variety of synthetic methods for
the preparation of bridgehead carbocyclic olefins have
been developed;6 however, relatively few methods for
anti-Bredt imines have appeared.6c,7 In most cases, anti-
Bredt imines have been generated by thermal and
photochemical decomposition of bridgehead azides;8 how-
ever, these methods are often complicated by the mixture
of rearranged isomers of bridgehead imines. Alterna-
tively, a few approaches involving Pb(OAc)4 oxidation of
an appropriate lactam9 and an intramolecular aza-Wittig
reaction10 have been appeared. Although nucleophilic
addition to these bridgehead imines with alcohols and
cyanide ion has been frequently carried out in trapping
experiments, no synthetic utilization with bridgehead
imines has been explored. Herein, we report the first
success in a facile, general entry to C-C bond formation
at bridgehead iminium ions and application in the
construction of 5-azatricyclo[6.3.1.01,5]dodecane (4).
The requisite tricyclic N,O-acetals 12-14 used in the
study were prepared as outlined in Scheme 1. Conden-
sation of 3-(2-bromoethyl)cyclohexanone ethylene ketal
(8)11 with 2-aminoethanol, 2-hydroxybenzylamine, and
2-(aminomethyl)benzyl alcohol (2 equiv were used in each
case) at heating with or without a solvent gave the
corresponding hydroxy amines 9-11. Deketalization
with HCl-MeOH (reflux) followed by heating in refluxing
CHCl3 or benzene resulted in the formation of the
conspicuous feature of 3 is an azatricyclic ring system
(shown by bold lines in 3) consisting of the combination
of the morphan and indolizidine nuclei sharing the
piperidine ring, namely, 5-azatricyclo[6.3.1.01,5]dodecane
(4). To our knowledge, such a unique ring system as well
as the structural feature of the morphan framework with
an alkyl substitution at the C-1 bridgehead position, with
the exception of only a single example3 for the latter case,
are not otherwise present in known natural and synthetic
compounds. Our interest in the synthetic study of 3 led
us to develop a facile, general formation of the morphans
possessing 1-alkyl substituents, i.e., 2, wherein R ) alkyl,
and the construction of the 5-azatricyclo[6.3.1.01,5]dodecane
core structure 4.
(6) For bridgehead olefin reviews, see: (a) Keese, R. Angew. Chem.,
Int. Ed. Engl. 1975, 14, 528. (b) Shea, K. J . Tetrahedron 1980, 36,
1683. (c) Warner, P. M. Chem. Rev. (Washington, D.C.) 1989, 89, 1067.
(d) Kraus, G. A.; Hon, Y.-S.; Thomas, P. J .; Laramay, S.; Liras, S.;
Hanson, J . Chem. Rev. (Washington, D.C.) 1989, 89, 1591. (e) Broden,
W. T. Tetrahedron 1989, 89, 1095.
(7) For a review of synthesis of azaadamantanes via bridgehead
imines, see: Eguchi, S.; Okano, T.; Takeuchi, H. Heterocycles 1987,
26, 3265.
We have recently been investigating4 the highly selec-
tive Lewis acid-induced nucleophilic alkylation with
(1) For a review of the synthesis of 2-azabicyclo[3.3.1]nonanes, see:
Bosch, J .; Bonjoch, J . Heterocycles 1980, 14, 505.
(8) See, for example: (a) Sasaki, T.; Eguchi, S.; Katada, T.; Hiroaki,
O. J . Org. Chem. 1977, 42, 3741. (b) Quast, H.; Eckert, P.; Seiferling,
B.; Peters, E.-M.; Perters, K.; von Schnering, H. G. Chem. Ber. 1985,
118, 3058. (c) Quast, H.; Eckert, P.; Seiferling, B. Chem. Ber. 1985,
118, 3535. (d) Radziszewski, J . G.; Downing, J . W.; Wentrup, C.;
Kaszynski, P.; J awdosiuk, M.; Kovacic, P.; Michl, J . J . Am. Chem. Soc.
1985, 107, 2799. (e) Wayne, G. S.; Snyder, G. J . J . Am. Chem. Soc.
1993, 115, 9860. See also ref 7.
(2) Sakamoto, K.; Tsujii, E.; Abe, F.; Nakanishi, T.; Yamashita, M.;
Shigematsu, N.; Izumi, S.; Okuhara, M. J . Antibiot. 1996, 49, 37.
(3) Cronyn, M. W.; Riesser, G. H. J . Am. Chem. Soc. 1953, 75, 1664.
(4) (a) Yamazaki, N.; Suzuki, H.; Aoyagi, S.; Kibayashi, C. Tetra-
hedron Lett. 1996, 37, 6161. (b) Yamazaki, N.; Kibayashi, C. Tetrahe-
dron Lett. 1997, 38, 4623.
(5) (a) Fawcett, F. S. Chem. Rev. (Washington D.C.) 1950, 47, 219.
(b) Wiseman, J . R.; Pletcher, W. A. J . Am. Chem. Soc. 1970, 92, 956.
(c) Ko¨brich, G. Angew. Chem., Int. Ed. Engl. 1973, 12, 464. (d)
Buchnan, G. L. Chem. Soc. Rev. 1974, 3, 41. (e) Maier, W. F.; Schleyer,
P. v. R. J . Am. Chem. Soc. 1981, 103, 1891.
(9) Toda, M.; Niwa, H.; Ienaga, K.; Hirata, Y. Tetrahedron Lett.
1972, 335.
(10) Sasaki, T.; Eguchi, S.; Okano, T. J . Am. Chem. Soc. 1983, 105,
5912.
(11) Mills, S. G.; Beak, P. J . Org. Chem. 1985, 50, 1216.
S0022-3263(97)01557-0 CCC: $14.00 © 1997 American Chemical Society

Communications
J . Org. Chem., Vol. 62, No. 24, 1997 8281
Ta ble 1. Nu cleop h ilic Alk yla tion w ith Or ga n om eta llic Rea gen ts on Cyclic N,O-Aceta ls 12-14
tricyclic N,O-acetals 12-14 (see Table 1) with construc-
tion of the morphan skeleton.
Sch em e 2
We then explored the possibility of introduction of an
alkyl group onto the bridgehead position of the morphan
core in the tricyclic N,O-acetals 12-14 with organome-
tallic reagents. With this aim we first employed 12 and,
according to our previous procedure for the formation of
“hydrazonium ions” from the hydrazine N,O-acetals,4 we
performed the substitution reaction using 2.5 equiv of
Et3Al as a Lewis acid with a nucleophilic alkyl group,
resulting in alkylation of the bridgehead position to afford
15a in 75% yield (Table 1, entry 1). When the allyl
Grignard reagent was employed without adding a Lewis
acid, only a trace of the 1-allylated product 15b was
observed (entry 2). This result suggested that the ability
of Lewis acid to induce the Grignard addition is necessary
to allow the initial formation of the iminium ion in
agreement with our previous observations.4b Thus, 12
was treated with Et2AlCl (2 equiv), which sufficiently
activates the N,O-acetal to cleave the C-O bond leading
to the iminium ion, and then the Grignard reagents (2
equiv), providing the corresponding 1-alkylated mor-
phans 15a -c in good yields (entries 2-5). This meth-
odology based on Lewis acid induced C-C bond formation
was successfully applied to 13 and 14, giving rise to the
1-alkylated morphans 16 (entries 6-8) and 17 (entries
9-12) in good to excellent isolated yields. The benzylic
substitution on the nitrogen in 16a and 16b could be
readily removed by hydrogenolysis over Pd-C to give the
2-alkylmorphans 2, wherein R ) Et (68% yield) and Pr
(70% yield), respectively.
The results of the bridgehead alkylation obtained
herein are consistent with an SN1 mechanism described
in Scheme 2. In this reaction Et2AlCl (or Et3Al) acts as
Lewis acid to weaken the C-O bond in 18, promoting
the formation of the reactive bridgehead iminium ion pair
to 19. Subsequent nucleophilic attack on 19 by the
Grignard reagents (or internal delivery4b of the ethyl
group via 19′ in the case where Et3Al was used) leads to
the 1-alkylmorphan 20, wherein C-C bond formation
proceeds with no stereochemical ambiguity with regard
to the newly created stereogenic center at C-1, since
attack on only one face is enforced by the rigid bicyclic
system.
Sch em e 3
With the required introduction of the alkyl group onto
the bridgehead carbon atom of the morphan framework
in hand, we proceeded to try to construct the fundamental
core ring system of FR901483 (3), 5-azatricyclo[6.3.1.01,5]-
dodecane (4). Thus, 13 was allowed to react with [2-(1,3-
dioxolan-2-yl)ethyl]magnesium bromide in the presence
of Et2AlCl in THF to give 21 in 75% yield (Scheme 3).
Hydrogenation of 21 over Pd-C was initially carried out
in MeOH to cleave the 2-hydroxybenzyl group and
continued after addition of 3 M HCl, resulting in deac-
etalization and subsequent in situ iminium ion cycliza-
tion and reduction to afford 4 in 71% overall yield.
In conclusion, we have demonstrated that the cyclic
N,O-acetals could serve as potential precursors for the
morphan ring construction and generation of reactive
bridgehead iminium ions via Lewis acid treatment, which
undergo alkylation with organometallics providing the
1-alkylated morphans. The methodology developed was
successfully applied to the synthesis of 5-azatricyclo-
[6.3.1.01,5]dodecane.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and analytical and spectral data for compounds 2 (R
) Et and Pr), 4, 9-14, 15a -c, 16a -c, 17a -c, and 21 (11
pages).
J O9715579