Bridgehead enolates: Substitution and asymmetric desymmetrization of small bridged carbonyl compounds by lithium amide bases

Article abstract of DOI:10.1021/ol034348l

(Matrix presented) Contrary to expectations, a number of bridged carbonyl compounds undergo facile bridgehead metalation with lithium amide bases. Diketone, lactone, lactam, and imide functions are all demonstrated to participate in this type of "bridgehe

Full text of DOI:10.1021/ol034348l

ORGANIC
LETTERS
2003
Vol. 5, No. 10
1673-1675
Bridgehead Enolates: Substitution and
Asymmetric Desymmetrization of Small
Bridged Carbonyl Compounds by
Lithium Amide Bases
,‡
Gerard M. P. Giblin,† Douglas T. Kirk,‡ Lee Mitchell,‡ and Nigel S. Simpkins*
GlaxoSmithKline, Department of Medicinal Chemistry, Neurology CEDD, The Frythe,
Hertfordshire AL6 9AR, U.K., and School of Chemistry, UniVersity of Nottingham,
UniVersity Park, Nottingham NG7 2RD, U.K.
nigel.simpkins@nottingham.ac.uk
Received February 27, 2003
ABSTRACT
Contrary to expectations, a number of bridged carbonyl compounds undergo facile bridgehead metalation with lithium amide bases. Diketone,
lactone, lactam, and imide functions are all demonstrated to participate in this type of “bridgehead enolate” chemistry, leading to a range of
substituted products. Meso compounds can also be desymmetrized in very high ee by asymmetric bridgehead metalation.
Metalation of ketones such as camphenilone 1 at the
bridgehead position is expected to be difficult or impossible
because the conventional enolate form of the resulting anion
would break Bredt’s rule (Figure 1).1 In fact, ketone 1 is
readily metalated by lithium tetramethylpiperidide (LTMP),
but the resulting “bridgehead enolate” (perhaps more ac-
curately described as an R-keto carbanion) undergoes such
rapid addition to the starting ketone that its interception by
alternative electrophiles has not been possible.2 A number
of other bridgehead metalations have also been described,
including carboxylation of imide 2 and aldol-type reaction
of diketopiperazine 3.3,4
While bridgehead metalation of small-bridge carbonyl
compounds is expected to be problematic, with larger systems
† GlaxoSmithKline.
‡ University of Nottingham.
(1) (a) Shea, K. J. Tetrahedron 1980, 36, 1683. (b) Warner, P. M. Chem.
ReV. 1989, 89, 1067. (c) Certain systems are known to undergo base-
catalysed bridgehead deuteration; see: Nickon, A.; Covey, D. F.; Huang,
F.; Kuo, Y.-N. J. Am. Chem. Soc. 1975, 97, 904.
(2) (a) Shiner, C. S.; Berks, A. H.; Fisher, A. M. J. Am. Chem. Soc.
1988, 110, 957. (b) The inability to trap bridgehead enolates has led to the
invention of indirect approaches; see, for example: Spitz, U. P.; Eaton, P.
E. Angew. Chem., Int. Ed. Engl. 1994, 33, 2220.
Figure 1. Carbonyl compounds known to undergo bridgehead
metalation.
10.1021/ol034348l CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/18/2003

a transition to “normal” enolate chemistry should be evident.5
Remarkably, ketone 4 undergoes kinetically controlled
enolate formation at the bridgehead position despite the
availability of an alternative methylene site for deprotona-
tion.6
LiCl as base (eq 1). In the diketone and imide cases (10 and
11), the desired product was accompanied by lesser amounts
of disilylated product (typically 10-20%) and unidentified
byproducts, which could be minimized by the use of LTMP
in place of LDA (eq 2).8 In the case of diketone 10, we
employed 2.5 equiv of base in the expectation that bridgehead
substitution might occur via a dianion9 although the reaction
most likely proceeds via initial formation of an enol silane.
In the next phase of exploration, we examined metalation
of lactam and imide compounds having shorter bridges,
Scheme 2 (eqs 3 and 4).
Recently, we demonstrated that bridgehead metalation-
substitution of ketones such as 5 is possible in high yield by
use of a lithium amide-in situ Me3SiCl quench protocol and
that enantioselective desymmetrization was possible using
a chiral lithium amide base.7 However, the scope of such
bridgehead metalations remains ill-defined, especially in
regard to interesting examples such as imide 2, which appear
to lie between the uncontrolled carbanion-like camphenilone
system and the well-behaved large bridge systems.
Here, we demonstrate that a range of bridged systems,
having small bridges (one to three atoms), with ketone, imide,
lactam, or lactone activating functions undergo lithium amide
mediated bridgehead metalation-substitution. We also show
that very high levels of enantioselectivity can be achieved
in asymmetric desymmetrization of bridged imides using the
chiral base method.
Scheme 2
The bicyclo[3.3.1]nonane systems 6, 7, 10, and 11, having
various carbonyl functions in one of the three-atom bridges,
provide interesting preliminary observations concerning the
viability of bridgehead substitution. Under our usual low-
temperature in situ quench conditions using Me3SiCl as the
electrophile, modest to good yields of the desired products
were obtained, Scheme 1 (eqs 1 and 2).
The bicyclo[3.2.1]octane lactam 14 underwent high-
yielding silylation using LDA-LiCl, and we found that the
use of sBuLi, as employed in metalations of the rather more
hindered imide 2, gave none of the desired product. Silylation
of the corresponding imide 15 proved more problematic, with
bis-silylation predominating. The unsaturated bicyclo[2.2.1]-
heptane lactam 18 gave no products of bridgehead substitu-
tion, but instead gave a high yield of silylated alkene 19.
When we removed the double bond from 18, either by
hydrogenation or by dihydroxylation-acetonide formation,
none of the desired mode of substitution could be achieved.
This system is related to the classical camphenilone example
1, and in line with previous work we were unable to intercept
the anion from this ketone, even using a large excess of Me3-
SiCl at low temperature. Thus, it seems that successful
metalations of these very small rigid systems still present a
problem.
Scheme 1
Asymmetric desymmetrization of the meso-imides 11 and
15 was carried out by use of chiral lithium amide base 20 or
the bis-lithiated base 21, Scheme 3.10,11
The use of (R,R)-bisphenylethylamide 20 enabled the
synthesis of (-)-13 in high yield and enantiomeric excess,
the process being considerably more efficient than the
corresponding reaction with LDA or LTMP. In the silylation
The lactone 6 and lactam 7 underwent surprisingly smooth
bridgehead silylation, using excess (1.2-1.8 equiv) LDA-
(3) Wanner, K. T.; Paintner, F. F. Liebigs Ann. 1996, 1941.
(4) (a) Yamaura, M.; Nakayama, T.; Hashimoto, H.; Shin, C.; Yoshimura,
J.; Kodama, H. J. Org. Chem. 1988, 53, 6035. (b) Williams, R. M.;
Armstrong, R. W.; Dung, J.-S. J. Am. Chem. Soc. 1984, 106, 5748. (c)
Eastwood, F. W.; Gunawardana, D.; Wernert, G. T. Aust. J. Chem. 1982,
35, 2289.
(5) (a) Khan, F. A.; Czerwonka, R.; Zimmer, R.; Reissig, H.-U. Synlett
1997, 995. (b) Magnus, P.; Parry, D.; Iliadis, T.; Eisenbeis, S. A.; Fairhurst,
R. A. J. Chem. Soc., Chem. Commun. 1994, 1543. (c) Wender, P. A.;
Mucciaro, T. P. J. Am. Chem. Soc. 1992, 114, 5878. (d) Rigby, J. H.; Moore,
T. L. J. Org. Chem. 1990, 55, 2959.
(7) Blake, A. J.; Giblin, G. M. P.; Kirk, D. T.; Simpkins, N. S.; Wilson,
C. Chem. Commun. 2001, 2668.
(8) In cases such as 6 or 7, excess base can be used to ensure smooth
conversion to product, whereas with substrates such as imide 11 the
possibility for double bridgehead substitution dictates the use of close to
stoichiometric amounts of base (usually 1.1 equiv). In typical metalations,
we added a solution of 1.1-1.2 equiv of base to a mixture of 1.0 equiv of
substrate and 3.0 equiv of electrophile in THF at -105 °C. The mixture
was then allowed to warm to room temperature before standard workup
and chromatography. Further details can be found in the Supporting
Information.
(6) Gwaltney, S. L., II; Sakata, S. T.; Shea, K. L. J. Org. Chem. 1996,
61, 7438.
(9) Berry, N. M.; Darey, M. C. P.; Harwood, L. M. Synthesis 1986, 476
and references therein.
1674
Org. Lett., Vol. 5, No. 10, 2003

ascertained the full scope of this procedure in terms of
substrate or electrophile, but we expect that similar substitu-
tions will be possible on other systems.
Scheme 3
The highly enantioselective silylation of imides 11 and
15 enables further selective transformations with synthetic
potential. First, it was interesting to note that silylimide (-)-
13 undergoes rather facile and high-yielding substitution at
the remaining bridgehead site, using the type of in situ
quenching procedure outlined above and with LTMP as base,
e.g., to give 27 and 28 (Figure 2).
of 15, base 20 proved less selective, providing (-)-17 in
70% ee. In this case we utilized the bis-lithium amide base
21,12 which then provided (+)-17 in 47% yield and 94% ee.
Although the Me3SiCl in situ quench procedure had
provided some remarkable new bridgehead silylation results,
we were interested in probing the possibilities for achieving
alternative bridgehead alkylation, acylation, etc. At present,
it appears that treatment of most of the aforementioned
substrates with lithium amide bases, followed by addition
of electrophiles in the conventional way, provides very low
levels of substitution product. However, by addition of chiral
base 20 to a mixture of imide 11 and an appropriate
electrophile, asymmetric C-alkylation or acylation is possible
(Table 1).13
Figure 2.
Second, the bridgehead silicon substituent in chiral imides
such as (-)-13 and (-)-17 exerts impressive control of the
regiochemistry of subsequent reactions of the imide function.
For example, completely regioselective reduction of these
compounds was possible using DIBAL in CH2Cl2 at -78
°C to give 29 and 30, which could be further reduced to 16
(88%) and 9 (83%), respectively, using Et3SiH and Me3-
SiOTf.14
Similarly, regioselective thionation to give 31 and 32 was
possible using Lawesson’s reagent,15 and again no minor
regioisomers could be detected. Although we have not
checked the ee of these lactam and thioimide products they
should correspond to the initial values achieved in the chiral
base reactions.
Table 1. Asymmetric Bridgehead Substitution of Imide 11
In conclusion, we have demonstrated the unexpectedly
wide scope of bridgehead substitution via lithium amide
metalation of carbonyl compounds having relatively short
bridges. The chiral lithium amide mediated desymmetriza-
tions of meso-imide substrates further adds to the repertoire
of these versatile reagents, and enables highly enantioselec-
tive access to certain bridged imides and lactams. Further
explorations of the scope and limitations of such bridgehead
metalations are ongoing, along with applications to bioactive
target molecules.
product
electrophile
E
yield (%) ee (%)
(-)-22 methyl iodide
(-)-23 allyl bromide
(-)-24 prenyl bromide
(-)-25 benzyl bromide
Me
57
42
50
52
56
97
95
98
95
98
CH2CHdCH2
CH2CHdC(Me)2
CH2Ph
(-)-26 pivaloyl chloride COtBu
Acknowledgment. We are grateful to the University of
Nottingham and GlaxoSmithKline for support of this work
under the CASE scheme and to the Engineering and Physical
Sciences Research Council (EPSRC) for a Fellowship to
L.M.
Although yields are somewhat modest at present, this being
in part due to bis-alkylation, in all cases the enantiomeric
excess of the product was excellent. We have not yet
Supporting Information Available: Typical procedures
for metalations, NMR data for all compounds, and HPLC
data for ee determinations. This material is available free of
charge via the Internet at http://pubs.acs.org.
OL034348L
(10) For previous chiral base desymmetrisation of imides, see: (a)
Adams, D. J.; Blake, A. J.; Cooke, P. A.; Gill, C. D.; Simpkins, N. S.
Tetrahedron 2002, 58, 4603. (b) Greenhalgh, D. A.; Simpkins, N. S. Synlett
2002, 2074.
(11) The absolute configurations of the chiral silylimides 13 and 17 were
determined by X-ray crystallography and the C-alkylated compounds are
assumed to belong to the same enantiomeric series; full details will be
published elsewhere.
(12) Bambridge, K.; Begley, M. J.; Simpkins, N. S. Tetrahedron Lett.
1994, 35, 3391.
(13) At this time, it is not clear why this procedure is required, and we
have been unable to satisfactorily monitor the course of the metalations
using deuterium-quenching experiments.
(14) For contributions to the area of regioselective imide reduction, see:
(a) Wijnberg, J. B. P. A.; Schoemaker, H. E.; Speckamp, W. N. Tetrahedron
1978, 34, 179. (b) Speckamp, W. N.; Hiemstra, H. Tetrahedron 1985, 41,
4367.
(15) Milewska, M. J.; Gdaniec, M.; Polonski, T. J. Org. Chem. 1997,
62, 1860.
Org. Lett., Vol. 5, No. 10, 2003
1675