A practical synthesis of S-quinuclidine-2-carboxylic acid and its enantiomer

Article abstract of DOI:10.1016/j.tetlet.2006.02.064

A short and convenient synthesis of (S)- and (R)-quinuclidine-2-carboxylic acids has been developed. The resolution of enantiomers has been accomplished by both chemical and enzymic means.

Full text of DOI:10.1016/j.tetlet.2006.02.064

Tetrahedron Letters 47 (2006) 2515–2516
A practical synthesis of S-quinuclidine-2-carboxylic acid
and its enantiomer
Yuan Mi and E. J. Corey*
Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street, Cambridge, MA 02138, USA
Received 26 January 2006; revised 8 February 2006; accepted 9 February 2006
Abstract—A short and convenient synthesis of (S)- and (R)-quinuclidine-2-carboxylic acids has been developed. The resolution of
enantiomers has been accomplished by both chemical and enzymic means.
Ó 2006 Elsevier Ltd. All rights reserved.
One of the least studied types of a-amino acids are
the bridgehead nitrogen bicyclic structures exemplified
by quinuclidine-2-carboxylic acid (1, S-enantiomer).
Although racemic 1 was first synthesized decades ago,1
it remains an obscure, little studied compound despite
its close structural relationship to the important
a-amino acids proline and pipecolinic acid. Very little
has been published on chiral 1.2 A simple and practical
synthesis of 1 could facilitate its study as an N-terminal
amino acid unit for synthetic peptides and other poten-
tial therapeutic agents,3 and as a building block for
enantioselective catalysts. Reported herein is a short
and effective synthesis of 1 and its enantiomer.
after extractive isolation (96%). Treatment of 3 in ether
solution with 2.5 equiv of aqueous NaCN at 0 °C fol-
lowed by dropwise addition of concentrated hydrochlo-
ric acid (DANGER, hydrogen cyanide generated, must
be carried out in a well ventilated hood) at 0 °C afforded
cyanohydrin 4 (99.4%) as a colorless oil after extractive
isolation.4 Reaction of 4 with 1.2 equiv of methanesulfo-
nyl chloride and 1.3 equiv of triethylamine in CH2Cl2
solution at ꢀ30 °C over 2 h gave cyanomesylate 5 as
an oil (>99%) after extractive isolation. The mesylate 5
was transformed into cyanoquinuclidine 6 by the
sequence: (1) N-deprotection with 5 equiv of CF3CO2H
in CH2Cl2 at 0 °C for 5 min and 23 °C for 1 h, (2) con-
centration under reduced pressure and reaction with
4 equiv of Et3N in CH3CN at reflux for 24 h, and (3)
extractive isolation, which provided 6 as a solid in
41% overall yield. The resolution of 6 was readily
accomplished using (+)-tartaric acid (1 equiv) in abso-
lute ethanol, which gave a crystalline salt as needles,
COO
N
H
H
1
and 2–3 further recrystallizations from methanol. Pure
The synthetic pathway to 1, which is summarized in
Scheme 1, started with commercial 4-(2-hydroxy-
ethyl)piperidine (2, Reilly Industries, Indianapolis, IN).
N-Acylation of 2 with 1 equiv of di-tert-butylpyrocar-
bonate (in 4:3 H2O–t-BuOH in the presence of 1.1 equiv
of NaOH) at 23 °C for 30 h provided the N-tert-butoxy-
carbonyl (Boc) derivative of 2 in >99% yield (100 g
scale). Swern oxidation of Boc-protected 2 (reaction
with dimethyl sulfoxide-oxalyl chloride reagent at
ꢀ60 °C for 30 min, addition of Et3N and warming to
ambient temperature) afforded the amide-aldehyde 3
23
6, mp 70–71 °C, ½aꢁD ꢀ88.6 (c 3.3, CHCl3) was obtained
from the salt by treatment with wet Ba(OH)2 or wet
NaHCO3 in CH2Cl2 and removal of solvent. The abso-
lute configuration of (ꢀ)-6 follows from the conversion
to (ꢀ)-1, the absolute configuration of which had previ-
ously been established.2 HPLC analysis using Chiral
Technologies AD column with 10% i-PrOH in hexane
for elution indicated an enantiomeric ratio of 98:2
(minor peak at 17.1 min, major peak at 19.3 min at
23 °C and a flow rate of 0.5 ml/min). From the mother
liquors (+)-6 was obtained via recrystallization of the
23
salt with (ꢀ)-tartaric acid as a colorless solid ½aꢁD
*
+84.2 (c 2.7, CHCl3) with an enantiomeric ratio of
Corresponding author. Tel.: +1 617 495 4033; fax: +1 617 495
0376; e-mail: corey@chemistry.harvard.edu
98.5:1.5 as determined by HPLC analysis.
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.02.064

2516
Y. Mi, E. J. Corey / Tetrahedron Letters 47 (2006) 2515–2516
OMs
CN
OH
CHO
OH
CN
MeSO2Cl
NaCN
1. Boc2O
2. Me2SO
ClCOCOCl;
Et3N
96%
HCl
H2O
99%
Et3N
99%
N
H
N
N
N
Boc
5
Boc
4
Boc
3
2
1. CF3CO2H,
CH2Cl2
2. Et3N
41%
1. (+)-Tartaric
acid, EtOH
Cl
1. ClCOCOCl
N
CO2Me
N
CO2H
N
6
CN
2. CH3OH
3. Ba(OH)2
2. H3O
H
7
1
HCl
Scheme 1.
i-Pr2O solution with Amano lipase PS and excess isopro-
penyl acetate at 0 °C for 11 days effected enantioselec-
tive O-acetylation to form 8. The crude product was
treated with CH3SO2Cl–Et3N in CH2Cl2 at ꢀ30 °C for
1 h to form a mixture of 8 and mesylate 9 as shown in
Scheme 2. Chromatography of this mixture on silica
OAc
CN
OH
CN
OH
CN
Amano
lipase PS
0 oC, 11 days
+
N
N
N
gel (EtOAc–hexane) provided the chiral acetate (ꢀ)-8
OAc
Boc
Boc
4
Boc
23
(46%, 92% of theory, 94% ee by HPLC analysis, ½aꢁD
(–)- 8
ꢀ31.6 (c 2.0, CHCl3)) and the chiral mesylate (+)-9
MsCl
Et3N
(42%, 84% of theoretical yield).
The chiral quinuclidine derivatives 6 and 7 can be used
to make not only a variety of useful amino alcohols
but also chiral diamines, compounds that are of interest
as chiral reagents. This is not conveniently accomplished
by the previously described routes (most of which lead
to racemic amino acid 11). The synthesis outlined in
Scheme 1 is short, simple, and easy to scale up. Two sim-
ple and efficient methods of resolution allow convenient
access to chiral material on a multigram scale. The best
literature method for obtaining chiral2 1 involves resolu-
tion of 2-hydroxymethylquinuclidine and a low yield
oxidation (aqueous KMnO4) to 1.2b
OMs
CN
N
Boc
(+)- 9
Scheme 2.
Hydrolysis of (ꢀ)-6 was effected by heating at reflux
with concentrated hydrochloric acid for 24 h to give
after evaporation to dryness under reduced pressure
References and notes
the hydrochloride of 1 (S-enantiomer) as a crystalline
1. (a) Prelog, V.; Cerkovnikov, E. Ann. Chem. 1937, 532, 83;
(b) Renk, E.; Grob, C. A. Helv. Chim. Acta 1954, 37, 2119–
2123; (c) Von Pracejus, H.; Kohl, G. Ann. Chem. 1969, 722,
1–11; (d) Bulacinski, A. B. Pol. J. Chem. 1978, 52, 2181–
2187.
2. (a) Rubstov, M. V.; Mikhlina, E. E. Zh. Obshch. Khim.
1955, 25, 2303–2310; (b) Langstrom, B. Chem. Scripta 1974,
5, 170–173.
3. See, for example, Bencherif, M.; Schmitt, J. D.; Bhatti, B.
S.; Crooks, P.; Caldwell, W. S.; Lovette, M. E.; Fowler, K.;
Reeves, L.; Lippiello, P. M. J. Pharm. Exp. Therapeut.
1998, 284, 886–894.
23
solid in 99% yield, ½aꢁD ꢀ39.7 (c 1.6, H2O).1 Treatment
of 1ÆHCl with oxalyl chloride in CH2Cl2 containing a
catalytic amount of dimethylformamide for 14 h gener-
ated the corresponding acid chloride, which by reaction
with methanol afforded the hydrochloride of 7. The
hydrochloride was dissolved in CH2Cl2 and stirred with
wet Ba(OH)2 at 23 °C for 2 min to give after filtration
23
and removal of solvent the methyl ester 7 (67%, ½aꢁD
ꢀ3.6 (c 0.5, CHCl3)).
Another efficient process for resolution has been devel-
oped using enantioselective enzyme-catalyzed acetyl-
ation of the racemic cyanohydrin 4. Treatment of 4 in
4. Excess cyanide in the aqueous layer after extraction can be
destroyed by the addition of aqueous NaOCl (commercial
bleach).