A new silicon lewis acid for highly enantioselective Mannich reactions of aliphatic ketone-derived hydrazones

Article abstract of DOI:10.1021/ja800830h

The first general method for the highly enantioselective Mannich reaction of aliphatic ketimines is reported. A new, second generation chiral silane Lewis acid has been developed that promotes the reaction between ketone-derived hydrazones and silyl ketene acetals, providing the β,β-disubstituted β-amino esters with good enantioselectivity even for the hydrazone derived from 2-butanone (methyl vs ethyl, 91% ee). Several examples are provided, including a reaction with a substituted (propanoate-derived) silyl ketene acetal. Copyright

Full text of DOI:10.1021/ja800830h

Published on Web 04/30/2008
A New Silicon Lewis Acid for Highly Enantioselective Mannich Reactions of
Aliphatic Ketone-Derived Hydrazones
Gregory T. Notte and James L. Leighton*
Department of Chemistry, Columbia UniVersity, New York, New York 10027
Received February 1, 2008; E-mail: leighton@chem.columbia.edu
Scheme 1
ꢀ-Amino acids have long been recognized as important building
blocks in natural products synthesis and medicinal chemistry1 and
oligomers thereof as peptidomimetics.2 As a direct result, massive
effort has in recent years been devoted to the development of
enantioselective Mannich reaction methodologies, and numerous
impressive successes have been recorded.3 With only very few
exceptions, however, these successes have been limited to the use
of aldimines as the electrophilic reaction component. Only a few
examples of highly enantioselective Mannich reactions with
Scheme 2
ketimines have been reported,4,5 and to our knowledge, there are
no reported examples of highly enantioselective reactions of
alipahtic ketone-derived imines. Due to the success of our strained
silane Lewis acid platform with aliphatic ketone-derived hydrazones
in other reactions,6 we decided to examine whether we could
address this important gap in the scope of enantioselective Mannich
reactions.
Phenylsilane 1 has been shown to be effective both for acylhy-
drazone Friedel-Crafts reactions7 and for acylhydrazone enol ether
[3 + 2] cycloaddition reactions,8 and it was hoped that this success
would extend to the addition of silyl ketene acetals (SKA) to
aliphatic ketone-derived hydrazones. Unfortunately, although 1 was
found to promote the reaction between hydrazone 2a and SKA 3,
Table 1. Highly Enantioselective Mannich Reactions with
Ketimines
the reaction gave 4a with only moderate efficiency and poor
enantioselectivity (Scheme 1).
Upon consideration of what changes could be readily made to
the silane Lewis acid to render it more effective, replacement of
the phenyl group with a more electronegative and sterically tunable
group seemed an attractive option (Scheme 2). Alkoxy groups fit
the bill on both counts and were also attractive in practical terms
due to the ready availability of alkoxytrichlorosilanes.9 Thus,
entry
R
timea
yield (%)
ee (%)
reaction of 5 and 6 with pseudoephedrine provided alkoxysilanes
7 and 8 in 70 and 95% yields, respectively, and as 2.5:1 and 2.2:1
mixtures of (unassigned) diastereomers, respectively. With these
new silanes in hand, our test reaction with hydrazone 2a was
reexamined, and gratifyingly, silanes 7 and 8 provided dramatically
improved performance, giving 4a in 81% and 74% ee, respectively.
Upon optimization, it was found that the use of PhCF3 as solvent
provided another boost in enantioselectivity, and under these
conditions (2a + silane in PhCF3, 23 °C, 30 min; add SKA 3, 30
min), silane 8 led to the production of 4a in 89% yield and 90%
ee. Given this performance, and its experimentally trivial, inex-
pensive, and high-yielding preparation, silane 8 was deemed most
effective.
With optimal conditions identified, we set out to examine the
scope of the reaction with respect to the hydrazone structure (Table
1). A range of aliphatic ketone-derived hydrazones 2a-f partici-
pated smoothly in the reaction, giving ꢀ,ꢀ-dialkyl-ꢀ-hydrazidoesters
4a-f in good yields and with high levels of enantioselection (entries
1-6). Remarkably, the hydrazone derived from 2-butanone (2c) is
a highly effective substrate, giving 4c in 91% ee (entry 3). To the
best of our knowledge, this is the highest enantioselectivity achieved
for a nucleophilic addition to a 2-butanone-derived imine.10 More
1
2
3
4
PhCH2CH2 (a)
PhCH2 (b)
Et (c)
CH2dCHCH2CH2 (d)
i-PrCH2CH2 (e)
TBSOCH2CH2 (f)
i-PrCH2 (g)
Cy (h)
30 min
30 min
30 min
30 min
30 min
30 min
22 h
89
89
84
74
83
66
74
71
60
90
96
91
89
92
91
96
97
88
5
6
7b
8c
9d
22 h
22 h
Ph (i)
a Refers to reaction time after addition of 3. b 3 equiv of 3, 1.6 equiv
of 8,
°C. c 4 equiv of 3, equiv of 8, in CH2Cl2. d Ar
p-NO2-C6H4, 10 °C.
0
2
)
highly substituted aliphatic substrates (2g,h) are well tolerated and
give superior enantioselectivities (entries 7 and 8), albeit in sluggish
reactions that require higher loadings of 8 and the SKA. Finally,
although it is not the focus of this study, we note that the
acetophenone-derived hydrazone 2i is an effective substrate, as well
(entry 9).
As one demonstration of the utility of this new and effective
method for aliphatic ketimine substrates, we have developed an
approach to an unusual 1,2,3,4-tetrahydroquinoline, a ring system
of importance in organic and medicinal chemistry.11 Thus, the
9
6676 J. AM. CHEM. SOC. 2008, 130, 6676–6677
10.1021/ja800830h CCC: $40.75
2008 American Chemical Society

C O M M U N I C A T I O N S
Scheme 3
Scheme 5
We have developed a new silane Lewis acid (8) that is effective
for the highly enantioselective Mannich reaction of SKAs with
aliphatic ketone-derived hydrazones. Silane 8 is trivial to prepare
on large scale at a nominal cost, and neither its preparation nor
use presents any significant environmental or toxicity issues.
Ongoing efforts will address both increased scope in this Mannich
reaction and a more detailed understanding of the mechanism.
Scheme 4
Acknowledgment. This work was supported by a grant from
the NSF (CHE-04-53853) and a focused Funding Award from
Johnson & Johnson. G.T.N. is the recipient of a postdoctoral
fellowship (F32GM080859) from the NIH. We thank Merck
Research Laboratories and Amgen for unrestricted support. We
thank Prof. G. F. R. Parkin and Mr. Kevin Yurkerwich for X-ray
analyses, performed with the support of NSF CHE-06-19638.
Mannich reaction of hydrazone 9 proceeded to give 10 in 86% yield
and 91% ee (Scheme 3). Hydrazide reduction with SmI212 gave 11
in 80% yield, and intramolecular Buchwald-Hartwig amination13
gave 1,2,3,4-tetrahydroquinoline 12 in 73% yield.
We are aware of only a few examples of asymmetric Mannich
reactions of substituted enolates with ketimines.4a,b,d SKA 13 (5:1
E:Z) was therefore prepared and employed in the Mannich reaction
(Scheme 4). Although the diastereoselectivity of the reaction was
low, we were gratified to find that the reaction proceed reasonably
efficiently (71% yield) and in 86% ee.
Supporting Information Available: Experimental procedures,
characterization data, and stereochemical proofs. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) EnantioselectiVe Synthesis of ꢀ-Amino Acids, 2nd ed.; Juaristi, E., Solos-
honok, V. A., Eds.; Wiley-VCH: New York, 2005.
(2) (a) Cheng, P. R.; Gellman, S. H.; DeGrado, W. F. Chem. ReV. 2001, 101,
3219. (b) Seebach, D.; Beck, A. K.; Bierbaum, D. J. Chem. BiodiV. 2004,
1, 1111.
As noted above, silane 8 is prepared and utilized as a 2.2:1
mixture of diastereomers. In addition, hydrazones 2 are synthesized
and utilized as ∼4-5:1 mixtures of E and Z isomers (except 2h,
which is a g20:1 mixture).14 Variation of the stoichiometry of 8
(0.5, 1.3, 2.0, and 4.0 equiv) under the otherwise standard conditions
for the reaction of hydrazone 2a led to only slight variations in the
enantioselectivity of the reactions (89, 90, 92, and 92% ee,
respectively). These data allow us to exclude the possibility that
the silane diastereomers react through separate pathways with
significantly different rates and enantioselectivities. While the yields
in Table 1 do not allow us to exclude the possibility that only the
major silane diastereomer is active, it will be noted that the 89%
yields obtained for 4a and 4b are exactly the theoretical maximum
for such a scenario. As for the hydrazone E and Z isomers, those
same 89% yields for 4a and 4b establish that both isomers must
be participating in the reaction. Indeed, when the reaction of 2a
was run with 2.0 equiv of 8 (and with only a 10 min complexation
time), 4a was isolated in 97% yield and 92% ee.
Taken together, these data suggest that silanes 8 and hydrazones
2 converge on a common complex, and that the reactions proceed
through a single common pathway.15 While we have thus far been
unsuccessful in our attempts to characterize a relevant silane-
hydrazone complex by X-ray crystallographic analysis, it will be
noted that we have done so previously with the complex derived
from (S,S)-1 and the benzoyl hydrazone of benzaldehyde.6 That
structure established (1) the convergence of both silane diastere-
omers on a single complex upon treatment with a hydrazone, and
(2) isomerization of the hydrazone trans isomer to cis in the
complex. On the basis of this precedent, it is reasonable to propose
the formation of a single complex 15 from silanes 8 and hydrazones
2 (Scheme 5). Attack of SKA 3 on the exposed front face of
complex 15—the back face being effectively blocked by the neo-
pentyl group—correctly predicts the observed sense of absolute
stereochemical induction.
(3) Enantioselective Mannich reactions have been extensively reviewed. See
ref 1 and: (a) Kobayashi, S.; Ishitani, H. Chem. ReV. 1999, 99, 1069. (b)
Liu, M.; Sibi, M. P. Tetrahedron 2002, 58, 7991. (c) Taggi, A. E.; Hafez,
A. M.; Lectka, T. Acc. Chem. Res. 2003, 36, 10. (d) Marques, M. M. B.
Angew. Chem., Int. Ed. 2006, 45, 348. (e) Friestad, G. K.; Mathies, A. K.
Tetrahedron 2007, 63, 2541. (f) Ting, A.; Schaus, S. E. Eur. J. Org. Chem.
2007, 5797. (g) Verkade, J. M. M.; van Hemert, L. J. C.; Quaedflieg,
P. J. L. M.; Rutjes, P. J. T. Chem. Soc. ReV. 2008, 37, 29.
(4) (a) Bravo, P.; Fustero, S.; Guidetti, M.; Volonterio, A.; Zanda, M. J. Org.
Chem. 1999, 64, 8731. (b) Tang, T. P.; Ellman, J. A. J. Org. Chem. 2002,
67, 7819. (c) Saaby, S.; Nakama, K.; Lie, M. A.; Hazell, R. G.; Jørgensen,
K. A. Chem.—Eur. J. 2003, 9, 6145. (d) Zhuang, W.; Saaby, S.; Jørgensen,
K. A. Angew. Chem., Int. Ed. 2004, 43, 4476. (e) Suto, Y.; Kanai, M.;
Shibasaki, M. J. Am. Chem. Soc. 2007, 129, 500.
(5) For a recent alternate approach to the synthesis of ꢀ,ꢀ-disubstituted ꢀ-amino
acids, see: (a) Minter, A. R.; Fuller, A. A.; Mapp, A. K. J. Am. Chem. Soc.
2003, 125, 6846. (b) Fuller, A. A.; Chen, B.; Minter, A. R.; Mapp, A. K.
J. Am. Chem. Soc. 2005, 127, 5376.
(6) Berger, R.; Duff, K.; Leighton, J. L. J. Am. Chem. Soc. 2004, 126, 5686.
(7) Shirakawa, S.; Berger, R.; Leighton, J. L. J. Am. Chem. Soc. 2005, 127,
2858.
(8) Shirakawa, S.; Lombardi, P. J.; Leighton, J. L. J. Am. Chem. Soc. 2005,
127, 9974.
(9) Masaoka, S.; Banno, T.; Ishikawa, M. J. Organomet. Chem. 2006, 182.
(10) Storer, R. I.; Carrera, D. E.; Ni, Y.; MacMillan, D. W. C. J. Am. Chem.
Soc. 2006, 128, 84.
(11) (a) Katritsky, A. R.; Rachwal, S.; Rachwal, B. Tetrahedron 1996, 52, 15031.
(b) Kouznetsov, V.; Palma, A.; Ewert, C.; Varlamov, A. J. Heterocycl.
Chem. 1998, 35, 761.
(12) Burk, M. J.; Martinez, J. P.; Feaster, J. E.; Cosford, N. Tetrahedron 1994,
50, 4399.
(13) (a) Wolfe, J. P.; Marcoux, J. F.; Buchwald, S. L. Acc. Chem. Res. 1998,
31, 805. (b) Hartwig, J. F. Acc. Chem. Res. 1998, 31, 852. (c) Mauger,
C. C.; Mignani, G. A. Aldrichimica Acta 2006, 39, 17.
(14) As noted by Burk (see ref 12), the 1H NMR spectra of ketone-derived
hydrazones are complicated by rotational isomerism. However, spectra taken
in CD3OD and DMSO-d6 give reasonably sharp spectra that reveal two
compounds (assigned as the E and Z isomers) and that do not change in
response to changes in temperature. See the Supporting Information.
(15) It is consistent with this proposal that in the reaction of 2a under the standard
conditions the product ee is constant throughout the reaction. See the
Supporting Information.
JA800830H
9
J. AM. CHEM. SOC. VOL. 130, NO. 21, 2008 6677