Cation-π control of regiochemistry of intramolecular Schmidt reactions en route to bridged bicyclic lactams

Article abstract of DOI:10.1021/ja068919r

The regiochemistry of the intramolecular Schmidt reaction of 2-substituted ketones has proved susceptible to control by the placement of an aromatic group at an adjacent position, permitting the selective formation of a series of bridged bicyclic lactams

Full text of DOI:10.1021/ja068919r

Published on Web 02/16/2007
Cation-π Control of Regiochemistry of Intramolecular Schmidt Reactions en
Route to Bridged Bicyclic Lactams
Lei Yao and Jeffrey Aube´*
Department of Medicinal Chemistry, UniVersity of Kansas, 1251 Wescoe Hall DriVe, Malott Hall, Room 4070,
Lawrence, Kansas 66045-7852
Received December 13, 2006; E-mail: jaube@ku.edu
Scheme 1
Bridged bicyclic lactams that incorporate the lactam nitrogen at
a bridgehead position have been of interest for many years because
they incorporate a “twisted amide” unable to achieve standard planar
geometry.1 Although much of the work devoted to these compounds
has focused on the hyperreactivity of the amide bond toward
hydrolysis, recent work by us2 and others3 suggests that there is
still much to be learned about the reactivity of these interesting
structures. Not surprisingly, the tendency of these compounds to
undergo rapid hydrolysis has complicated their synthesis. Although
standard methods for amide bond formation have been used, such
routes often proceed in poor yields or are accompanied by difficulty
in product isolation.1 In this paper, we describe a solution to the
problem of bridged bicyclic lactam synthesis that utilizes the acid-
promoted reaction of an alkyl azide and a ketone4 in which the
regiochemistry is controlled by a through-space interaction of an
aryl group with a cationic leaving group in a key reaction
intermediate.
Scheme 2
The utility of the intramolecular Schmidt reaction to the problem
of bridged lactam was recently reported by Stoltz and Tani (Scheme
1a).5 Using 3-(azidoethyl)cyclopentanone, it proved possible to form
a mixture of two lactams from which the desired quinuclidone 1a
was isolated by recrystallization. This sequence is noteworthy
because it permitted the isolation and characterization of the iconic
bridged lactam 1a for the first time and also demonstrated the loss
of N2 as a powerful driving force for the formation of this unstable
ring system.
The placement of the azide-containing side chain at a carbon
nonadjacent to the ketone means that only bridged lactams are
possible in this reaction. For the analogous R-substituted ketones,
however, two regiochemical outcomes can be envisioned (Scheme
1b). Such reactions almost always afford fused lactam by formal
insertion of the azide into the proximal C-C bond of the reactive
ketone (path a).4 We set out to obtain a new class of bridged
lactams2 that would be formed by the regiochemical alternative ring
expansion via path b.
The first example of an intramolecular Schmidt reaction to afford
such a bridged bicyclic lactam was encountered unexpectedly
(Scheme 2).6,7 Thus, triene 2 underwent an intramolecular Diels-
Alder reaction to afford a ketone that further reacted to give a
mixture of the expected fused lactam 3a and a bridged isomer 3b.
We propose that the regiochemistry of the Schmidt reaction involves
the selective migration of the C-C bond antiperiplanar to the
leaving N2+ substituent. In this scenario, chairlike cyclohexanones
can only afford a bridged lactam when the azide-containing side
chain occupies a pseudoaxial orientation (see the Supporting
Information for details of this published argument4). Once this
condition is satisfied, then antiperiplanar migration of the C-C
bond anti to the equatorial N2+ group in A would afford the fused
isomers, whereas the axially disposed leaving group in B is
necessary for the formation of the bridged isomers.
We first attempted to affect this equilibrium by disfavoring
isomer A through the placement of a bulky group at R2. To this
end, the isopropyl-containing isomer 4 was synthesized and
submitted to Lewis acid treatment. In this case, essentially the same
ratio of bridged/fused adducts was obtained. Conversely, the
analogue bearing a phenyl group at the same position gave a
dramatically different result. In this case, fused isomer 7a was
formed exclusively in 85% yield, making this reaction the most
efficient example of this sequence observed to date. Although the
phenyl and isopropyl have similar A values,8 we wondered if
apparently exclusive intermediacy of A might be explained by an
attractive through-space interaction between the positively charged
leaving group and the phenyl group. Although commonly invoked
in small-molecule/protein binding interactions,9 such cation-π
interactions have only occasionally been proposed in stereoselective
synthesis.10 In one case, we proposed such an interaction as a
controlling feature of certain asymmetric Schmidt reactions.11
9
2766
J. AM. CHEM. SOC. 2007, 129, 2766-2767
10.1021/ja068919r CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Synthesis of Fused and Bridged Lactams
(although it very likely exists in the ground state). The effect of
phenyl group placement in compounds 8b-d is depicted in the
bottom part of Scheme 3.
Overall, these results indicate an important role for stereoelec-
tronic effects in controlling the regiochemistry of this reaction. The
fact that the vast majority of 2-azidoalkyl ketones undergo
rearrangement to afford fused lactams may reflect both the
conformation of the reactive intermediate azidohydrin and the
stability of normal vs twisted lactams. In particular, the observation
of a small amount of twisted amide from compound 8b indicates
that appropriately constrained compounds can lead to bridged
isomers, if only as minor products. However, the observation of
through-space control of these intramolecular Schmidt reactions is
especially provocative. This laboratory will carry out further
investigations of this interesting effect in this chemistry and in other
contexts as well.
yield (%)
entry
8
−
10
R1
R2
9
10
1
2
3
4
a
b
c
H
Ph
H
Ph
96
57
20
10
0
17
51
65
t-Bu
t-Bu
t-Bu
d
p-(MeO)C6H4
Scheme 3
Acknowledgment. We thank the National Institutes of Health
Institute of General Medical Sciences (GM-49093) for financial
support, and Christopher Katz and Michal Szostak for experimental
assistance.
Supporting Information Available: Additional discussion of the
effect of the orientation of the C-2 substituent on regiochemistry,
experimental details, and characterization data for new compounds. This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) For twisted amide reviews: see (a) Greenberg, A. Mol. Struct. Energ.
1988, 7, 139-178. (b) Greenberg, A., Breneman, C. M., Liebman, J. F.,
Eds. Amide Linkage: Selected Structural Aspects in Chemistry, Biochem-
istry, and Materials Science; Wiley: New York, 2000.
(2) Yao, L.; Wrobleski, A. D.; Golden, J. E.; Powell, D. R.; Aube´, J. J. Am.
Chem. Soc. 2005, 127, 4552-4553.
(3) For some recent papers regarding the reactivity of twisted amides: see
(a) Kirby, A. J.; Komarov, I. V.; Wothers, P. D.; Feeder, N. Angew. Chem.,
Int. Ed. 1998, 37, 785-786. (b) Kirby, A. J.; Komarov, I. V.; Feeder, N.
J. Chem. Soc., Perkin Trans 2 2001, 522-529. (c) Bashore, C. G.;
Samardjiev, I. J.; Bordner, J.; Coe, J. W. J. Am. Chem. Soc. 2003, 125,
3268-3272.
Accordingly, we hypothesized that an appropriately situated
aromatic group might be able to stabilize isomer B and provide a
direct route to bridged isomers.
(4) (a) Aube´, J.; Milligan, G. L. J. Am. Chem. Soc. 1991, 113, 8965-8966.
(b) Milligan, G. L.; Mossman, C. J.; Aube´, J. J. Am. Chem. Soc. 1995,
117, 10449-10459.
Control experiments demonstrated that neither placement of a
phenyl group at the R carbon (entry 1) nor the axial orientation of
the side chain (entry 2) alone is sufficient to steer the reaction
toward fused lactam product (although the small amounts of 10b
observed are consistent with the results seen in Scheme 212).
HoweVer, the combination of an axial azide-containing tether and
an aromatic group in a 1,3-diaxial relationship with the leaVing
(5) Tani, K.; Stoltz, B. M. Nature 2006, 441, 731-734.
(6) Golden, J. E.; Aube´, J. Angew. Chem., Int. Ed. 2002, 41, 4316-4318.
(7) The first example of any bridged product from an intramolecular Schmidt
reaction actually occurred during synthetic studies toward aspidospermi-
dine. However, this involved the reaction of a nonadjacent azidoalkyl chain
with a ketal, affording a bridged orthoaminal product and not a lactam.
See: (a) Iyengar, R.; Schildknegt, K.; Aube´, J. Org. Lett. 2000, 2, 1625-
1627. (b) Iyengar, R.; Schildknegt, K.; Morton, M.; Aube´, J. J. Org. Chem.
2005, 70, 10645-10652.
+
N2 group in the intermediate azidohydrin diVerts the reaction so
that bridged product predominates. The fact that the ratio of bridged
to fused isomers increases with a more electron-rich aromatic group
provides strong support that a 1,3 cation-π effect is in play (cf.
entries 3 and 4).
The relevant intermediates are shown in Scheme 3. Our
hypothesis is supported by several observations. First, the fact that
8a affords only fused product 9a rules out any effect arising from
differences in intrinsic migration aptitudes. In this case, the
thermodynamically favored conformation of 8a affords a reactive
intermediate that can only lead to 9a. Interestingly, the alternative
conformation bearing an axial side chain, that could in principle
lead to bridged isomer, does not appear to react in this example
(8) The A values for Ph and i-Pr are 2.8 and 2.21, respectively.
(9) (a) Dougherty, D. A. Science 1996, 271, 163-168. (b) Ma, J. C.;
Dougherty, D. A. Chem. ReV. 1997, 97, 1303-1324.
(10) (a) Lakshminarasimhan, P.; Sunoj, R. B.; Chandrasekhar, J.; Ramamurthy,
V. J. Am. Chem. Soc. 2000, 122, 4815-4816. (b) Stratakis, M.; Froudakis,
G. Org. Lett. 2000, 2, 1369-1372. (c) Yamada, S.; Morita, C. J. Am.
Chem. Soc. 2002, 124, 8184-8185. (d) Yamada, S.; Saitoh, M.; Misono,
T. Tetrahedron Lett. 2002, 43, 5853-5857. (e) Kaanumalle, L. S.;
Sivaguru, J.; Arunkumar, N.; Karthikeyan, S.; Ramamurthy, V. Chem.
Commun. 2003, 116-117.
(11) Katz, C. E.; Aube´, J. J. Am. Chem. Soc. 2003, 125, 13948-13949.
(12) Interestingly, this same reaction when carried out using TiCl4 provided
compound 9b as a single product in 92% yield, suggesting that the
regiochemistry of the reaction is Lewis acid-dependent. Further work to
examine this point is in progress.
JA068919R
9
J. AM. CHEM. SOC. VOL. 129, NO. 10, 2007 2767