Highly enantioselective synthesis of substituted piperidines using the chiral lithium amide base approach

Article abstract of DOI:10.1055/s-1999-2813

The symmetry-breaking enolisation reaction of a meso-piperidine diester using a chiral bis-lithium amide base allows access to alkylated derivatives in highly diastereo- and enantioselective fashion (> 98% ee).

Full text of DOI:10.1055/s-1999-2813

1292
LETTER
Highly Enantioselective Synthesis of Substituted Piperidines using the Chiral
Lithium Amide Base Approach
Nicholas J. Goldspink, Nigel S. Simpkins, Marion Beckmann
A School of Chemistry, The University of Nottingham, University Park, Nottingham, NG7 2RD, UK
Fax (0115) 9513564; E-mail: Nigel.Simpkins@Nottingham.ac.uk
BAgrEvo UK Limited, Chesterford Park, Saffron Walden, Essex, CB10 1XL, UK
Received 10 June 1999
Abstract: The symmetry-breaking enolisation reaction of a meso-
Ph
N
Ph 5
H
H
H
SiMe3
O
piperidine diester using a chiral bis-lithium amide base allows ac-
cess to alkylated derivatives in highly diastereo- and enantioselec-
tive fashion (≥ 98% ee).
Li
LiCl, Me3SiCl
O
O
O
N
N
ca. -100 °C,
Ph
Ph
Key words: chiral lithium amide, asymmetric synthesis
6 80% (95% ee)
4
Scheme 1
Introduction
Over recent years there have been significant develop-
ments in the application of chiral lithium amide base reac-
tions to asymmetric synthesis.1 Typically, such reactions
involve a symmetry-breaking enolisation or metallation
process to give an enantiomerically enriched nucleophilic
intermediate, which can then be reacted with electrophiles
to give useful non-racemic products. Most of this chemis-
try involves enolisation of cyclic ketones, e.g. the synthe-
sis of enol silane 1,2 but more recently we have applied the
same principle to other systems, for example to generate
chiral organometallics such as 2,3 or the chiral phos-
pholane oxide 3.4
Here we describe the application of this symmetry-break-
ing concept to a very different situation, which has excit-
ing implications for the synthesis of chiral heterocyclic
systems.
Since a number of symmetrical heteroaromatic com-
pounds are readily available it is a simple matter to access
the corresponding saturated meso derivatives via highly
stereoselective hydrogenation. In our initial studies, we
have applied this approach to dipicolinic acid 7, as shown
in Scheme 2.
(i) dimethoxypropane
HCl, MeOH, 75%
OSiMe3
Me
(ii) H2, Pd/C, 200 psi
MeO2C
N
CO2Me
Me
Ph
HO2C
N
CO2H
EtOH, 91%
(iii) PhCH2Br, Na2CO3
O
Bn
Ph
P
Ph
O
Cr(CO)3
CH2Cl2, H2O, 86%
tBu
7
8
1 (ca. 90% ee)
2 (99% ee)
3 (ca. 90% ee)
Ph
Ph
9
N
N
Li
Li
Ph
Ph
Most such enantioselective deprotonation reactions de-
scribed to date involve discrimination between enan-
tiotopic hydrogens activated by a single common
functional group. We expected that a considerable broad-
ening in the scope for applications of chiral lithium
amides would be possible if multifunctional substrates
could be employed. In this type of chiral base reaction the
acidic hydrogens would be activated by separate func-
tional groups.
MeO2C
N
CO2Me
THF, -78 °C
E
Bn
electrophile
10
Scheme 2
Transformation of diacid 7 into a suitable symmetrical di-
ester 8 was carried out very straightforwardly as shown.6
Several related derivatives were prepared, having differ-
ent nitrogen protecting groups, but since the N–benzyl
compound 8 behaved well in initial deprotonation reac-
tions, all subsequent work was carried out with this series.
We recently described a specific example of this concept,
involving the symmetry-breaking reactions of various
carbocyclic systems having a ring-fused imide, e.g. con-
version of the cyclopropane 4 into silylated product 6, me-
diated by chiral lithium amide 5, Scheme 1.5
Initial chiral base studies with the simple lithium amide
base 5 gave rather poor results, and it was not until we
turned to the use of the bis-lithium amide base 9 that we
Synlett 1999, No. 8, 1292–1294 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Highly Enantioselective Synthesis of Substituted Piperidines using the Chiral Lithium Amide Base
1293
achieved useful results. To our delight, treatment of 8 with
this base, followed by alkylation with benzyl bromide,
furnished product 10a in good yield as a single diastere-
omer, and in ≥98% ee.7,8 Similar excellent levels of dias-
tereocontrol and enantiocontrol were seen with a number
of other electrophiles, as indicated in the Table.9
9, THF, -78 °C
then MeOH
MeO2C
MeO2C
N
CO2Me
N
CO2Me
Bn
Bn
11 30% ( 89% ee)
8
(i) LiAlH4, THF
Table Asymmetric Alkylations of Diester 8
BnOCH2
N
CH2OBn
(ii) NaH, BnBr
Bu4N+ I-
Bn
12 51% overall
Scheme 3
rial.14 Clearly, this finding has implications for the synthe-
sis of this type of trans-substituted piperidine (and
perhaps other types of heterocycle), which have important
applications as chiral auxiliaries,15 as well as being com-
ponents of natural products.16 To date, we have not yet ex-
plored the use of alternative protonating sources, which
might allow a more stereoselective access to 11, but this
is clearly an attractive possibility.
Aldol reactions with various carbonyl compounds have
also been attempted, but give less consistent results, and
to date we have not secured the stereochemical assign-
ments.10
A key aspect of the new enantioselective substitution
chemistry of 8 is that it should allow further regioselective
modification of the product ester groups. We anticipated
that this would be possible in simple examples by subse-
quent reaction at the unsubstituted, and therefore less hin-
dered, ester position. Starting with compound 10a,
Scheme 4 illustrates how this is possible for reduction us-
ing DIBAL-H, and for hydrolysis under typical basic con-
ditions.
Although the sense of enantioselectivity in the reaction of
diester 8 with bis-lithium amide 9 is not readily rationa-
lised, we can propose several explanations for the high de-
gree of diastereoselectivity in the alkylations of the
intermediate enolate, Figure.
Bn
N
A
N
B
MeO2C
MeO2C
OMe
OMe
Bn
LiO
DIBAL-H
LiO
CH2Cl2, -78 °C
N
CO2Me
Ph
OH Bn
13 74%
MeO2C
N
CO2Me
Ph
Figure
Bn
NaOH
10a
If the exocyclic enolate has an equatorial disposition of
both the neutral amine group and the N-benzyl group then
selectivity may be determined mainly by the stereoelec-
tronic effect of the ring nitrogen as shown in structure A
(i.e. alkylation anti to the lone pair).11 Alternatively, if A1,3
strain considerations resulted in population of the N-in-
vertomer B, then the axial benzyl group provides an obvi-
ous source of facial shielding.12
MeOH, H2O
HO2C
N
CO2Me
Ph
Bn
14 70%
Scheme 4
Clearly, further transformation of these systems is plausi-
ble, including the formation of additional rings, making
them attractive starting points for natural product synthe-
sis.
During the course of the alkylation studies we found that
enolisation of 8, followed simply by re-protonation by ad-
dition of MeOH, returned meso diester 8, along with the
corresponding C2-symmetric diastereomer 11, Scheme The new chiral base chemistry described above opens up
3.13
new opportunities for the synthesis of chiral piperidines in
a highly stereocontrolled fashion. We expect that analo-
gous possibilities exist for other ring sizes and for rings in-
corporating different heteroatoms and substitution
patterns, and we are actively pursuing these avenues.
The latter compound, although recovered in only 30 %
yield from this process, proved to have an ee of 89%. Con-
version of this diester into the known bis-benzyl ether 12
served to confirm the absolute configuration of this mate-
Synlett 1999, No. 8, 1292–1294 ISSN 0936-5214 © Thieme Stuttgart · New York

1294
N. J. Goldspink et al.
LETTER
3.72(1H, d, J 12.5, CHHPh), 3.77 (3H, s, CO2Me), 3.80 (1H,
d,J 16.2, NCHHPh), 4.90 (1H, d, J 16.2, NCHHPh), 7.11(2H,
d, J 8.1), 7.18–7.28 (4H, m) and 7.32–7.39 (4H, m); dC(67.5
MHz; CDCl3) 18.0 (CH2), 28.2 (CH2), 33.5 (CH2), 46.8(CH2),
50.8 (CH3), 51.1 (CH3), 52.5 (CH2), 58.2 (CH), 63.0(C), 126.6
(CH), 126.7 (CH), 127.7 (CH), 128.0 (CH), 128.3(CH), 130.3
(CH), 136.1 (C), 140.8 (C), 173.9 (C=O) and 174.0 (C=O); m/
z (Found: M+H+ 382.2011.C23H27NO4 requires M+H,
382.2019).
The enantiomeric excess was determined by HPLC analysis
using a Chiralcel OD column with 1% i-PrOH in hexane as
eluant, using UV detection at 254 nm. Retention times were
15.1 min (minor) and 16.5 min (major).
Acknowledgement
We are grateful to The University of Nottingham and to AgrEvo UK
Limited for support of N.J.G.
References and Notes
(1) For reviews, see (a) O’Brien, P. J. Chem. Soc., Perkin
Trans. 1 1998, 1439. (b) Cox, P. J.; Simpkins, N. S.
Tetrahedron: Asymmetry 1991, 2, 1. (c) Simpkins, N. S.
Advances in Asymmetric Synthesis Stephenson, G. R. (Ed.),
Blackie Academic, 1996. (d) Koga, K. Pure Appl. Chem.
1994, 66,1487.
(8) The relative and absolute configuration shown for 10 has been
proved by X-ray crystallography for 10b, full details will be
published elsewhere.
(2) For the most recent studies of this reaction, see Chatani,
H.;Nakajima, M.; Kawasaki, K.; Koga, K. Heterocycles
1997,46, 53.
(9) The enantiomeric excess has been measured for 10a, 10c and
10e, the other examples are expected to have the same ee.
(10) Reaction with benzaldehyde gave three aldol products in a
roughly 2:1:2 ratio, whereas the corresponding reactions
involving either acetaldehyde or isobutyraldehyde gave single
diastereomeric aldolproducts (after chromatography) in
modest yield (ca. 40%). Thestereochemistry of these aldol
products has not been assigned.
(11) See Nagumo, S.; Mizukami, M.; Akutso, N.; Nishida, A.;
Kawahara, N. Tetrahedron Lett. 1999, 40, 3209, and
references therein.
(12) Seebach, D.; Sting, A. R.; Hoffmann, M. Angew. Chem.
Int.Ed. Engl. 1996, 35, 2708.
(3) Ewin, R. A.; Price, D. A.; Simpkins, N. S.; MacLeod, A.
M.;Watt, A. P. J. Chem. Soc., Perkin Trans. 1 1997, 401.
(4) Hume, S. C.; Simpkins, N. S. J. Org. Chem. 1998, 63, 912.
(5) Adams, D. J.; Simpkins, N. S.; Smith, T. J. N. Chem.Commun.
1998, 1605.
(6) Chenevert, R.; Dickman, M. J. Org. Chem. 1996, 61, 3332.
(7) A solution of bis-lithium amide 9 was prepared by
dropwiseaddition of a solution of BuLi (0.79 mmol) to the
appropriatediamine (173 mg, 0.41 mmol) in THF (4 ml) at
-78 °C. Theresulting red coloured solution was warmed
briefly to r.t. (20 min) before cooling to -78 °C and addition of
a solution of 8 (100 mg, 0.34 mmol) in THF (4 mL). The
mixturewas stirred at -78 °C for 1 h before addition of benzyl
bromide (1mL). After a further 1h at -78 °C the mixture was
allowed to warm to r.t. overnight, before quenching with
NaHCO3 (5 mL) and extraction of the product into Et2O (3 x
10 mL). Theorganic extract was dried (MgSO4), and
evaporation of the solvent under reduced pressure gave the
crude product, which waspurified by flash column
chromatography on silica-gel (10% Et2O in petroleum ether)
to give the product 10a as an off-white solid.Recrystallisation
from petroleum ether then gave colourlesscrystals of 10a (101
mg, 78%), mp 122–123 °C. Data for 10a
(13) Solladie-Cavallo, A.; Bouchet, M. J.; Wermouth, C. G. Org.
Mag. Res. 1983, 21, 367.
(14) Najdi, S.; Kurth, M. J. Tetrahedron Lett. 1990, 31, 3279.
(15)For a review, see (a) Whitesell, K. Chem. Rev. 1989, 89,1581.
For a recent example of the use of such C2-symmetric
pyrrolidines, see (b) Shi, M.; Satoh, Y.; Makihara, T.; Masaki,
Y.Tetrahedron: Asymmetry 1995, 6, 2109.
(16) For a recent review, see Bailey, P. D.; Millwood, P. A.; Smith,
P. D. Chem Comm. 1998, 633.
[a]D +27 (c 1.7, CHCl3); dH (400 MHz; CDCl3) 1.41–1.53
(2H,m), 1.68–1.76 (2H, m), 1.90–1.98 (2H, m), 2.66 (1H, d,
J12.5, CHHPh), 3.58 (1H, m, CHN), 3.63 (3H, CO2Me),
Article Identifier:
1437-2096,E;1999,0,08,1292,1294,ftx,en;L07699ST.pdf
Synlett 1999, No. 8, 1292–1294 ISSN 0936-5214 © Thieme Stuttgart · New York