Asymmetric aza-Wittig reactions: Enantioselective synthesis of β-quaternary azacycles

Article abstract of DOI:10.1002/anie.200601383

A hindrance no more: The direct asymmetric synthesis of nitrogen heterocycles containing β-quaternary stereocenters, widely found for example in alkaloids, is highly challenging. The novel approach shown exploits the desymmetrization of prochiral diketo azides to unmask the asymmetric quaternary center, and represents the first reported examples of the asymmetric aza-Wittig reaction. (Chemical Equation Presented).

Full text of DOI:10.1002/anie.200601383

Communications
introduced from simple prochiral 1,3-dicarbonyl precursors 1
bearing an amine equivalent by the desymmetrizing forma-
tion of a keto imine 2 (Scheme 1).[2]
Scheme 1. Desymmetrizing imine formation by amine equivalent
“NH2” for the construction of quaternary asymmetric centers.
Bonjoch and co-workers have demonstrated that a
diastereoselective variant of this strategy is possible by
using a-chiral amines in their studies on reductive amination
of 2-(2-oxooethyl)cycloalkyl 1,3-diones.[3,4] A conceptually
much more powerful and general strategy would employ
external chiral reagents in place of nonrecyclable chiral
amines and further would leave the imine intact for subse-
quent derivatization. Given the rapid and reversible nature of
imine formation, the potential for reagent-based acceleration
of the reaction and/or retention of the stereochemical
integrity of the products that are formed seems low. The
aza-Wittig reaction of iminophosphoranes with carbonyl
compounds,[5,6] however, is an irreversible imine-forming
reaction and also allows for the introduction of external
chirality through the use of chiral ligands on phosphorus.
Thus, the Staudinger reaction of azidodiketones 3 with an
appropriate chiral phosphorus reagent 4 would generate a
chiral iminophosphorane, which undergoes selective meta-
thesis with one of the two (now diastereotopic) carbonyl
groups to yield, irreversibly, enantioenriched 2 and the
corresponding phosphorus(V) oxide (Scheme 2). The anal-
Asymmetric Synthesis
DOI: 10.1002/anie.200601383
Asymmetric Aza-Wittig Reactions:
Enantioselective Synthesis of b-Quaternary
Azacycles**
Duanpen Lertpibulpanya, Stephen P. Marsden,*
Ignacio Rodriguez-Garcia, and Colin A. Kilner
Polysubstituted nitrogen heterocycles are prevalent in phar-
maceuticals and biologically important natural product tar-
gets, and new approaches to these systems are in constant
demand. A frequently occurring motif is the presence of an
asymmetric all-carbon quaternary center in the 3-position of
pyrrolidines, piperidines, and their polycyclic derivatives. All-
carbon quaternary stereocenters are amongst the most
challenging constructs in modern synthesis,[1] and a new
approach to such functionality would be of considerable
utility. It occurred to us that this moiety could potentially be
Scheme 2. Asymmetric aza-Wittig reactions mediated by chiral
phosphorus(III) reagents 4.
[*] Dr. S. P. Marsden, C. A. Kilner
School of Chemistry
ogies with asymmetric variants of the carbon-based Wittig
reaction are clear,[7,8] but to our knowledge there have been
no reports of asymmetric variants of the aza-Wittig reaction.
Herein we outline the successful demonstration of the first
examples of this process.
The azido-1,3-diketone substrates 3a–d were prepared in
four steps commencing from either pentane-1,3-dione or
cyclohexane-1,3-dione as shown in Scheme 3. For the chiral
phosphorus(III) reagents, we chose the known and readily
available proline-derived diazaphospholidine 4a[9] and oxa-
zaphospholidine 4b,[10] and the cyclohexyldiamine-derived
diazaphospholidine 4c (Scheme 3).[11]
University of Leeds
Leeds, LS29JT (UK)
Fax: (+44)113-343-6425
E-mail: s.p.marsden@leeds.ac.uk
D. Lertpibulpanya, Dr. I. Rodriguez-Garcia
Department of Chemistry
Imperial College London
London SW72AY (UK)
[**] We thank the Royal Thai Government for Ph.D. support (D.L.) and
the University of Almeria, Spain for research leave (I.R.-G.).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5000
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 5000 –5002

Angewandte
Chemie
mized conditions were applied to the range of substrates 3a–
d, and the results are shown in Table 1.
The reproducibility of the reaction was verified by
carrying out duplicate runs—the ee values shown are the
Table 1: Desymmetrization of diketo azides 3a–d to yield 6a–d.[a]
Scheme 3. Synthesis of substrates 3a–d. Reagents and conditions:
a) aqueous NaOH, MeI or BnBr; b) 1m NaOH, allyl bromide, Bu4NI,
room temperature; c) Cy2BH, THF, 08C, then NaOAc, I2; d) N aN,
Bu4NI, aqueous acetone, room temperature. Overall yields: 3a 44%,
3b 41%, 3c 29%, 3d 18%. Bn=benzyl; Cy=cyclohexyl.
3
Entry
Azide
Reagent
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
7
8
3a
3a
3b
3b
3c
3c
3d
3d
4b
4c
4b
4c
4b
4c
4b
4c
67
74
69
71
75
78
92
95
43.0
29.5
60.5
24.5
29.0
43.0
38.5
57.0
Initial studies with diketo azide 3a and diazaphospholi-
dine 4a showed that the reaction proceeded to conversion
after about 56 hours at room temperature. Initially we
attempted to assay the enantiomeric purity of the keto
imine product from the crude reaction mixture by chiral shift
NMR spectroscopic studies using BINOL as an additive.
These studies showed that significant asymmetric induction
was occurring (up to ca. 57% ee), but the values were found
not to be reproducible between runs of identical experiments.
We suspected that trace moisture was catalyzing racemization
through imine hydrolysis to the achiral aminodiketone, and
the viability of this pathway was verified by monitoring the
ee value of a sample of the crude imine which was left exposed
to atmospheric moisture. The enantiomeric purity decreased
steadily from 50–60% to 0% over a period of three days. We
therefore repeated the experiments with rigorous exclusion of
water and further trapped the crude keto imine product as the
stable N-methanesulfonyl enamine 5 by treatment with
methanesulfonyl chloride and triethylamine (Scheme 4).
[a] Reactions carried out in Et2O or THF, at room temperature or elevated
temperature; for individual reaction conditions, see the Supporting
Information. [b] Yield of isolated purified product (average of two runs).
[c] Average of two runs (all values Æ2%) as determined by chiral HPLC
analysis on Chiralcel OD or OJ columns. Ac=acetyl.
average of the two runs with a maximum variance of Æ 2%. In
all cases the yields of the isolated products were good to
excellent. In general the oxazaphospholidine 4b gave higher
asymmetric induction for the acyclic substrates whereas the
diazaphospholidine 4c gave higher asymmetric induction for
the cyclic substrates. The maximum levels of asymmetric
induction were around 60% ee. Though this value is not yet at
the high levels observed in many modern asymmetric trans-
formations, this result represents the first successful demon-
stration that the asymmetric aza-Wittig reaction is a viable
process. We anticipate that further tuning of the reagents and/
or substrates will lead to enhanced enantioselectivity.
The absolute sense of asymmetric induction was deter-
mined for keto enamide 6d. We exploited the differentiated
reactivity of the ketone and enamine by carrying out chemo-
selective reduction of the ketone with sodium borohydride to
afford 7 (Scheme 5). Derivatization of the resulting equatorial
Scheme 4. Trapping of the initially formed keto imines leads to
preservation of the enantiomeric integrity of the reaction. Ms=metha-
nesulfonyl.
alcohol with (1S)-camphanic chloride gave ester
8
(Scheme 5). The material that was formed had 58% de by
1H NMR spectroscopy and HPLC, thus confirming the
Upon chiral GC analysis of 5 we were delighted to find that
the enantiomeric purity of the crude imine had been retained
in the isolated product. It should be noted that the trapping
strategy not only safeguards the enantiomeric integrity of the
products but also further differentiates the two desymme-
trized functional groups—the remaining ketone is electro-
philic whereas the newly formed enamine is nucleophilic.
Encouraged by these results, we further optimized the
reaction by a) utilizing the more reactive phosphorus(III)
reagents 4b and 4c to shorten the reaction times and
b) changing the trapping regime from mesylation to the
higher-yielding and more reliable N-acetylation. These opti-
Scheme 5. Chemoselective manipulation of keto enamides and proof
of absolute stereochemistry. Reagents and conditions: a) NaBH4, room
temperature (93% yield); b) (1S)-camphanic chloride, DMAP, DCE,
reflux (98% yield). DMAP=4-(dimethylamino)pyridine; DCE=1,2-
dichloroethane.
Angew. Chem. Int. Ed. 2006, 45, 5000 –5002
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5001

Communications
Bosch, Angew. Chem. 1999, 111, 408 – 410; Angew. Chem. Int.
Ed. 1999, 38, 395 – 397; d) D. SolØ, J. Bonjoch, S. García-Rubio,
E. Peidró, J. Bosch, Chem. Eur. J. 2000, 6, 655 – 665.
measured asymmetric induction. The major diastereoisomer
was isolated by preparative HPLC, and crystals were grown
for X-ray diffraction analysis.[12] The obtained crystal struc-
ture confirmed that the iminophosphorane had attacked the
pro-R carbonyl group of 3d.
In summary, we have disclosed the first example of a new
class of asymmetric transformation, the asymmetric aza-
Wittig reaction, in the context of the desymmetrization of
prochiral azido-1,3-diketones. Further work to improve the
levels of enantioselectivity, to extend the range of substrates
for desymmetrization, and to apply this reaction in target
synthesis is in progress.
[5] For reviews of aza-Wittig processes, see: a) Y. G. Gololobov,
L. F. Kashukin, Tetrahedron 1992, 48, 1353 – 1406; b) A. W.
Johnson, Ylides and Imines of Phosphorus, Wiley, New York,
1993; c) S. Eguchi, Y. Matsushita, K. Yamashita, Org. Prep.
Proced. Int. 1992, 24, 209 – 243; d) P. M. Fresneda, P. Molina,
Synlett 2004, 1 – 17; e) S. Eguchi, ARKIVOC 2005, 2, 98 – 119.
[6] Synthesis of tetrahydropyridines by aza-Wittig reactions: a) M.
Pailer, E. Haslinger, Monatsh. Chem. 1970, 101, 508 – 511;
b) P. H. Lambert, M. Vaultier, R. Carrie, J. Chem. Soc. Chem.
Commun. 1982, 1224 – 1225; c) M. Vaultier, P. H. Lambert, R.
Carrie, Bull. Soc. Chim. Belg. 1985, 94, 449 – 456; d) M. Vaultier,
P. H. Lambert, R. Carrie, Bull. Soc. Chim. Fr. 1986, 83 – 92.
[7] For a review of asymmetric Wittig processes, see: T. Rein, T. M.
Pedersen, Synthesis 2002, 579 – 594.
Experimental Section
[8] For intramolecular desymmetrizing Wittig-type reactions of
diketones, see: a) B. M. Trost, D. P. Curran, J. Am. Chem. Soc.
1980, 102, 5699 – 5700; b) B. M. Trost, D. P. Curran, Tetrahedron
Lett. 1981, 22, 4929 – 4932; c) T. Mandai, Y. Kaihara, J. Tsuji, J.
Org. Chem. 1994, 59, 5847 – 5849; d) A. V. Bedekar, T. Wata-
nabe, K. Tanaka, K. Fuji, Tetrahedron: Asymmetry 2002, 13,
721 – 727; e) J. Yamazaki, A. V. Bedekar, T. Watanabe, K.
Tanaka, J. Watanabe, K. Fuji, Tetrahedron: Asymmetry 2002,
13, 729 – 734.
[9] J. M. Brunel, O. Legrand, S. Reymond, G. Buono, J. Am. Chem.
Soc. 1999, 121, 5807 – 5808.
[10] W. J. Richter, Chem. Ber. 1984, 117, 2328 – 2336.
[11] D. Smyth, H. Tye, C. Eldred, N. W. Alcock, M. Wills, J. Chem.
Soc. Perkin Trans. 1 2001, 2840 – 2849.
[12] CCDC-603132 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
A thoroughly flame-dried two-necked flask, fitted with a dry reflux
condenser and connected to a Schlenk line through a rotaflo stopcock,
was charged with an azidodiketone substrate 3a–d (0.5 mmol,
1.0 equiv), and the apparatus was evacuated and then purged with a
positive pressure of nitrogen. Freshly distilled dry solvent (see the
Supporting Information for individual experimental details) was
added to the flask by a septum, and a solution of the phosphane 4b or
4c (0.6 mmol, 1.2 equiv) was added, the total volume of solvent being
5 mL. The septum was replaced with a glass stopper, and the reaction
was allowed to proceed either at room temperature or with heating
(see the Supporting Information). At the end of the reaction, solvent
was removed in vacuo to give the crude imine. The apparatus was
then purged with a positive pressure of nitrogen, the glass stopper was
replaced with a rubber septum, and DCE (5 mL), NEt3 (4.0 equiv),
Ac2O (2 equiv), and DMAP (0.1 equiv) were successively added to
the flask. The septum was replaced with a glass stopper, and the
mixture was heated in an oil bath at 85–908C for 5–9 h. The mixture
was then cooled to room temperature, diluted with dichloromethane
(40 mL), washed with water (50 mL) and brine (50 mL), and dried
with MgSO4. The solvent was evaporated to give a residue, which was
preadsorbed on silica gel and purified by flash column chromatog-
raphy.
Received: April 7, 2006
Published online: July 3, 2006
Keywords: asymmetric synthesis · aza-Wittig reaction · azides ·
.
enantioselectivity · phosphanes
[1] For reviews, see: a) J. Christoffers, A. Baro, Quaternary Stereo-
centers. Challenges and Solutions for Organic Synthesis, VCH,
Weinheim, 2005; b) I. Denissova, L. Barriault, Tetrahedron 2003,
59, 10105 – 10146; c) J. Christoffers, A. Mann, Angew. Chem.
2001, 113, 4725 – 4732; Angew. Chem. Int. Ed. 2001, 40, 4591 –
4597; d) E. J. Corey, A. Guzman-Perez, Angew. Chem. 1998, 110,
402 – 415; Angew. Chem. Int. Ed. 1998, 37, 388 – 401; e) K. Fuji,
Chem. Rev. 1993, 93, 2037 – 2066.
[2] For reviews of desymmetrization processes in asymmetric syn-
thesis, see: M. C. Willis, J. Chem. Soc. Perkin Trans. 1 1999,
1765 – 1784.
[3] a) D. SolØ, J. Bonjoch, Tetrahedron Lett. 1991, 32, 5183 – 5186;
b) D. SolØ, J. Bosch, J. Bonjoch, Tetrahedron 1996, 52, 4013 –
4028; c) J. Bonjoch , D. SolØ, R. Carrillo, E. Peidró, J. Bosch,
Tetrahedron 2001, 57, 6011 – 6017.
[4] For applications in total synthesis, see: a) D. SolØ, J. Bonjoch, J.
Bosch, J. Org. Chem. 1996, 61, 4194 – 4195; b) J. Bonjoch, D.
SolØ, S. García-Rubio, J. Bosch, J. Am. Chem. Soc. 1997, 119,
7230 – 7240; c) D. SolØ, J. Bonjoch, S. García-Rubio, E. Peidró, J.
5002 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 5000 –5002