Asymmetric synthesis of (S)-4-oxopipecolic acid by a 3+3 atom-unit assembly strategy

Article abstract of DOI:10.1002/ejoc.200390028

(S)-4-Oxopipecolic acid has been prepared in enantiomerically pure form from two readily accessible starting materials, (R)-N-(cyanomethyl)-4-phenyloxazolidine (2) and 2-(methoxymethoxy)allyl chloride, (3, MOM-allyl chloride). This strategy exploits the complementary electrophile/nucleophile reactivity at each end of this pair of 3-atom unit building blocks. Thus, alkylation of the anion of 2 with 3 (creation of the piperidine C-2-C-3 bond) followed by Lewis acid induced cyclisation (creation of the C-5-C-6 bond), leads principally to a bicyclic intermediate 5 which incorporates a protected form of the title compound. Subsequent chemical transformations lead to the target molecule in an overall yield of 20% for the five-step sequence. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejoc.200390028

FULL PAPER
Asymmetric Synthesis of (S)-4-Oxopipecolic Acid by a 3؉3 Atom-Unit
Assembly Strategy
Karolin Partogyan-Halim,[a] Laure Besson,[a] David J. Aitken,*[a,b] and
Henri-Philippe Husson[a]
Keywords: Nitrogen heterocycles / Asymmetric synthesis / Cyclisation / Amino acids
(S)-4-Oxopipecolic acid has been prepared in enantiomer- duced cyclisation (creation of the C-5−C-6 bond), leads prin-
ically pure form from two readily accessible starting mat-
erials, (R)-N-(cyanomethyl)-4-phenyloxazolidine (2) and 2-
(methoxymethoxy)allyl chloride, (3, MOM-allyl chloride).
This strategy exploits the complementary electrophile/nucle-
ophile reactivity at each end of this pair of 3-atom unit build-
cipally to a bicyclic intermediate 5 which incorporates a pro-
tected form of the title compound. Subsequent chemical
transformations lead to the target molecule in an overall yield
of 20% for the five-step sequence.
ing blocks. Thus, alkylation of the anion of 2 with 3 (creation ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
of the piperidine C-2−C-3 bond) followed by Lewis acid in-
Germany, 2003)
Introduction
acyclic building blocks derived from the chiral pool.[11Ϫ14]
However, satisfactory stereocontrolled construction of the
4-oxopipecolate skeleton remains a contemporary chal-
lenge.
4-Oxopipecolic acid (1) is an unusual non-proteinogenic
α-amino acid, present as its (S)-enantiomer in some natur-
ally occurring cyclic depsipeptide antibiotics isolated from
streptomyces strains.[1] Recently, it has also been used as an
intermediate in the synthesis of protein tyrosine phosphat-
ase modulators,[2] NMDA receptor antagonists,[3] and
thrombin inhibitors.[4] As a consequence of its increasing
biological interest and its close structural and synthetic re-
lationships with the equally important 4-hydroxypipecolic
acid[5] and other ring-substituted derivatives,[6] a number of
synthetic strategies have been investigated for the ra-
cemic[7,8] and enantioselective[8Ϫ15] preparation of 1 in free
or protected form. Some preparations of enantiomerically
pure 1 have been achieved through chemical or enzymatic
resolution of derivatives of the racemate.[8,9] Asymmetric
synthesis strategies have included the use of nonracemic
chiral auxiliaries in olefin/iminium cyclisations,[3b] nucleo-
philic additions to 1-acylpyridinium salts,[5a] [2ϩ3] cycload-
ditions between methylenecyclopropane and a nitrone fol-
lowed by thermal rearrangement of the resulting isoxazolid-
ine,[10] or the construction of the piperidine ring from
We became interested in an unusual route to 1, based on
a 3ϩ3 atom-unit assembly. This is not an obvious retrosyn-
thetic analysis of pipecolic acid derivatives,[15] or indeed
piperidines in general,[16] whether the planned synthesis be
enantioselective or not; however, the approach seemed par-
ticularly well adapted for 1, as Scheme 1 illustrates. Differ-
ential nucleophile/electrophile activity on either side of a
chiral nitrogen synthon Ϫ effectively a chiral azomethine
ylide Ϫ required complementary reactivity for each of the
α-carbon atoms of an acetone equivalent. Previous work
from our laboratory on chiral nonracemic N-(cyano-
methyl)oxazolidines[17] suggested that compound 2 would
be a useful candidate for the former fragment,[18] while an
enol ether of an α-haloacetone appeared to be a convenient
synthetic equivalent of the latter fragment. Encouraged by
some preliminary studies on the use of such reagents for
entry to piperidine derivatives,[19Ϫ21] we undertook the
asymmetric synthesis of (S)-4-oxopipecolic acid [(S)-1]
according to this strategy.
[a]
´
´
Laboratoire de Chimie Therapeutique, UMR 8638 associee au
`
´
´
´
CNRS et a l’Universite Rene Descartes, Faculte des Sciences
Pharmaceutiques et Biologiques,
4 avenue de l’Observatoire, 75270 Paris cedex 06, France
Fax: (internat.) ϩ 33/(0)143291403
E-mail: husson@pharmacie.univ-paris5.fr
[b]
´
Laboratoire SEESIB Ϫ CNRS UMR 6504, Departement de
´
Chimie, Universite Blaise Pascal Ϫ Clermont-Ferrand II,
`
24 avenue des Landais 63177 Aubiere cedex, France
Fax: (internat.) ϩ 33/(0)473407717
E-mail: aitken@chisg1.univ-bpclermont.fr
Scheme 1
268
 2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/03/0102Ϫ0268 $ 20.00ϩ.50/0 Eur. J. Org. Chem. 2003, 268Ϫ273

Asymmetric Synthesis of (S)-4-Oxopipecolic Acid by a 3ϩ3 Atom-Unit Assembly Strategy
FULL PAPER
simple treatment with LDA/HMPA in THF at low temper-
ature followed by a proton quench with ammonium chlor-
ide solution, with virtually quantitative recovery
(Scheme 2). Thus, 4b may be considered as material for
potential recycling in this synthesis.
Results and Discussion
A summary of the synthesis is presented in Scheme 2.
After some reflection, we selected 2-(methoxymethoxy)allyl
chloride (MOM-allyl chloride, 3) as the acetone equivalent
in the synthesis. Reagent 3 was prepared conveniently in
high yield by using the literature procedure.[22] Depro-
tonation of N-(cyanomethyl)-4-phenyloxazolidine (2)[23]
with LDA/HMPA at low temperature, and treatment of the
resulting anion with 3, gave a mixture of the two diastereo-
isomers 4a and 4b in a 71:29 ratio with an overall yield
of 79%. All known previous alkylations of 2 by using this
procedure[18,23Ϫ25] were known to give preferentially the (S)
configuration at the new chiral centre and we assumed, for
the time being, that this would also be the case for the ma-
jor isomer 4a. There was some circumstantial support for
Cyclisation of 4a required the generation of an iminium
ion from the oxazolidine system, for which an oxophilic
Lewis acid such as BF3·Et2O seemed appropriate. In the
event, following low-temperature treatment of 4a with this
reagent, two piperidine-containing structures 5 and 6 were
obtained, in yields which varied depending on the condi-
tions. These results are presented in Table 1. The trans-
formation was sluggish and starting material 4a was par-
tially recovered if the Lewis acid was not present in at least
stoichiometric amounts, while a significant excess of the
reagent increased the formation of 6. Optimum formation
of 5 was observed with 1.5 equiv. of the Lewis acid, and
was slightly better in diethyl ether than in THF.
1
this assumption, in that the H NMR spectrum of 4a dis-
played a vicinal coupling constant of 2.5 Hz for the non-
equivalent oxazolidine ring C-2 protons; this phenomenon
had previously been observed for (S)-alkylated derivatives
of 2, while (R) isomers generally display a coupling con-
stant of around 4.5 Hz.[23,24] Major isomer 4a therefore pos-
sessed the correct configuration for the target molecule, and
was separated from 4b by standard chromatography before
being used in the next step. In this respect, compound 3
was a more useful acetone equivalent than 2-methoxyallyl
bromide, which leads to inseparable diastereomeric al-
kylation products in the alkylation of 2.[19]
Table 1. Lewis acid mediated cyclisation of compound 4a
Equivs. BF3·Et2O
Solvent
Yield of 5 [%]
Yield of 6 [%]
0.5
1.5
2.5
1.5
THF
THF
THF
Et2O
36
72
38
84
12
26
62
10
We interpret the competition between formation of 5 and
Although we did not exploit the minor isomer 4b further 6 in terms of the interactions shown in Scheme 3. BF3-in-
in this work, it is interesting to note that it could be epi- duced oxazolidine ring opening to give an iminium com-
merised to an approximately 50:50 mixture of 4a/4b by pound is followed smoothly by piperidine ring closure,
Scheme 2
Eur. J. Org. Chem. 2003, 268Ϫ273
269

K. Partogyan-Halim, L. Besson, D. J. Aitken, H.-P. Husson
Mineral acid mediated hydrolysis of the acetal function
FULL PAPER
which creates an electrophilic centre at C-4; this is conveni-
ently trapped by the Lewis acid complexed oxygen atom of unseparated 5a/5b mixtures was unfortunately accom-
(path a) to give the acetal system 5. Alternatively, the Lewis panied by facile elimination of HCN, to give a good yield
acid complexed oxygen atom may attack the electrophilic of the unstable (and unuseful) cyclic enaminone 7 (Scheme
acetal carbon atom of the MOM group, leading to a trans- 2). To avoid this, it was decided to convert the nitrile into a
fer of this acetal function to the phenylethoxy side chain carboxyl function prior to acetal hydrolysis. Basic hydrogen
(path b). This latter route becomes more favourable in the peroxide solution effected the smooth transformation of 5a/
presence of excess Lewis acid which has an activating effect 5b into a mixture of amide epimers 8. The product obtained
at the MOM centre. Two diastereomers of 5 were formed in this way was sufficiently clean to be used directly in the
in a 1:1 ratio, as a result of nonselective ring closure to next step, which was fortunate, since, remarkably, one of
form the cyclic acetal, giving a mixture of epimers at the the two stereoisomers could not be eluted from a silica gel
bridgehead carbon atom (C-4 of the piperidine system). column. Successive base and then acid treatment of amide
The isomers could be separated by chromatography, and epimer mixture 8 completed the transformation of the carb-
1
their H NMR spectroscopic data are consistent with the oxylate and liberated the 4-oxo function; the alcohol group
structures 5a and 5b (Scheme 3). Structural analysis of these freed in this process underwent spontaneous lactonisation
compounds was aided by examination of Dreiding models. with the carboxylic acid, finally giving a single bicyclic
In particular, the 2-H signal is shifted upfield in 5a com- product 9 in 72% yield. The presence of the chiral benzylic
pared with that of 5b (δ ϭ 3.96 and 4.21 ppm, respectively) amine type centre was the guarantee that 9 was enantio-
due to its proximity to the anisotropic cone of the phenyl merically pure and retained the (S) configuration at the C-
group in the former isomer, while the inverse applies to the 2 centre. On one occasion, we detected a very small amount
signal of 6-H(eq) (δ ϭ 3.19 and 2.90 ppm, for 5a and 5b, (about 3%) of the epimerised compound, epi-9, which was
respectively). Long-range W coupling (J ϭ 2.0 Hz) was ob- easily removed by chromatography.
served between 2-H and 6-H(eq) for 5a only. These observa-
The final step required removal of the chiral auxiliary to
tions are consistent with complete retention of the (S) con- free the amino acid system, an operation which was
figuration at the piperidine C-2 chiral centre.
achieved conveniently by hydrogenolysis of 9 in aqueous
acid solution in the presence of palladium on charcoal. Tar-
get molecule (S)-1 was thus obtained in 70% yield after pas-
sage through an ion-exchange resin, and has spectral and
physicochemical properties identical with those indicated in
the literature, notably [α]2D4 ϭ Ϫ17 (c ϭ 1.0, H2O) {ref.[8]
[α]2D5 ϭ Ϫ17 (c ϭ 1.0, H2O)}. In both the solid state and in
aqueous solution, (S)-1 exists essentially as its hydrate 10,
as evidenced by NMR, IR spectroscopic and mass spectro-
metric data. This propensity for addition reactions at the
C-4 ketone explains the need to carry out the final hydro-
genolysis step in non-alcoholic aqueous solution; in the
presence of methanol or ethanol as co-solvent, hemiacetal
and acetal adducts are obtained.
In summary, this original 3ϩ3 route compares well with
previous asymmetric syntheses and gives the title com-
pound reproducibly as a single enantiomer in just over 20%
overall yield from the two readily available starting mat-
erials 2 and 3. This figure does not take into account poten-
tial recycling of materials, notably the minor diastereomer
4b. Since C-2- and C-α-substituted derivatives of compound
Scheme 3
Both 5 and 6 were final products under the reaction con- 2 are accessible,[26] this strategy may turn out to be useful
dition, i.e. no Lewis acid catalysed equilibration between for the preparation of C-2- and/or C-6-substituted derivat-
the two structures was observed in control reactions. Since ives of the title product.
it seemed possible to influence the formation of one or the
other structures from the above observations, and since
Experimental Section
both were potential candidates for continuation through to
the target molecule 1, we pondered for a while on which
intermediate to favour. The decision came down principally
to arguments of stability, for compound 6 was difficult to
obtain pure and underwent degradation easily. We therefore
focused our efforts on the bicyclic derivative 5, despite the
fact that an extra chiral centre had been temporarily created
in this intermediate.
General Remarks: Melting points were recorded with a Kofler hot-
plate and are uncorrected. Optical rotations were obtained using a
PerkinϪElmer 141 MC polarimeter. NMR spectra were recorded
1
with a Bruker AC-300 spectrometer operating at 300 MHz for H
and at 75 MHz for 13C; chemical shifts (δ) are given in ppm and
were measured with reference to residual solvent peaks, except for
spectra recorded in D2O, in which case dioxane was added as an
270
Eur. J. Org. Chem. 2003, 268Ϫ273

Asymmetric Synthesis of (S)-4-Oxopipecolic Acid by a 3ϩ3 Atom-Unit Assembly Strategy
FULL PAPER
internal standard. IR spectra were obtained with a PerkinϪElmer
1600 Fourier transform instrument. Chemical ionisation mass spec-
tra were recorded with a Nermag 10-10 instrument with ammonia
as the vector gas. TLC was carried out using Merck silica gel 60
F254 aluminium foil plates (0.2 mm thickness); components were
detected first under UV light then by staining with ethanolic phos-
phomolybdic acid solution. Preparative flash chromatography was
layers were dried with MgSO4 and the solvents evaporated under
reduced pressure. Flash chromatography of the residual oil with
EtOAc/cyclohexane (9:1) gave facile separation of a diastereomeric
mixture 5a/5b (ratio 1:1; 3.70 g, 84%) and 6 (0.44 g, 10%). In gen-
eral, characterisation/identification of 5 was done on the diastere-
omeric mixture, which was used as such in the next step; on one
occasion, the stereoisomers were separated by repeated flash chro-
carried out with 200Ϫ400 mesh silica gel (Merck Kieselgel 60 or matography for more detailed NMR analysis.
SDS silica 60 ACC). Diethyl ether and THF were distilled under
Mixture 5a/5b: Colourless oil. MS: m/z ϭ 289 [MH]ϩ. IR (film):
argon from sodium benzophenone ketyl. Diisopropylamine and
HMPA were distilled under argon from CaH2. Compounds 2[23]
and 3[22] were prepared in high yields and purity according to liter-
ature procedures without any significant modifications. All other
reagents and solvents were used as obtained from commercial
sources without further purification.
ν˜ ϭ 2242 cmϪ1. C16H20N2O3 (288.3): calcd. C 66.65, H 6.99, N
9.72; found C 66.93, H 7.11, N 9.52. Fast Isomer 5a: Rf ϭ 0.40
1
(EtOAc/cyclohexane, 3:7). H NMR (CDCl3): δ ϭ 2.04 (ddd, J ϭ
14.3, 10.1, 4.5 Hz, 1 H), 2.11 (dd, J ϭ 14.5, 5.5 Hz, 1 H), 2.40
(dddd, J ϭ 14.1, 10.5, 5.3, 3.7 Hz, 1 H), 2.64 (ddd, J ϭ 14.4, 10.7,
3.6 Hz, 1 H), 3.19 (dddd, J ϭ 15.1, 10.6, 4.4, 2.1 Hz, 1 H), 3.37 (s,
3 H), 3.70 (ddd, J ϭ 15.3, 10.1, 5.4 Hz, 1 H), 3.86 (dd, J ϭ 13.3,
7.9 Hz, 1 H), 3.96 (ddd, J ϭ 10.7, 5.5, 2.0 Hz, 1 H), 4.18 (dd, J ϭ
Enol Ether 4: Under argon, a 2.2  solution of n-butyllithium in
hexane (13.4 mL, 29.5 mmol) was added to a solution of diisoprop-
ylamine (3.90 mL, 29.5 mmol) in THF (20 mL) at Ϫ40 °C. After 7.8, 5.1 Hz, 1 H), 4.36 (dd, J ϭ 13.3, 5.1 Hz, 1 H), 4.75 (d, J ϭ
stirring at this temperature for 15 min, the solution was cooled to 7.3 Hz, 1 H), 4.93 (d, J ϭ 7.3 Hz, 1 H), 7.21Ϫ7.48 (m, 5 H) ppm.
Ϫ70 °C and HMPA (15.1 mL, 98.4 mmol) was added, followed by 13C NMR (CDCl3): δ ϭ 34.4 (CH2), 37.2 (CH2), 42.4 (CH), 45.1
the dropwise addition of a solution of compound 2 (4.62 g,
24.6 mmol) in THF (10 mL). After 30 min, neat MOM-allyl chlor-
ide (3; 5.04 g, 36.9 mmol) was added dropwise. The mixture was
stirred at Ϫ70 °C for a further 3 h, then a saturated aqueous solu-
(CH2), 55.7 (CH3), 65.3 (CH2), 67.4 (CH), 89.9 (CH2), 100.2 (Cq),
120.2 (Cq), 126.8 (CH), 127.7 (CH), 128.8 (CH), 138.0 (Cq) ppm.
Slow Isomer 5b: Rf ϭ 0.35 (EtOAc/cyclohexane, 3:7). 1H NMR
(CDCl3): δ ϭ 1.81 (ddd, J ϭ 15.6, 10.9, 5.1 Hz, 1 H), 2.31 (dddd,
tion of NH4Cl (100 mL) was added and the mixture allowed to J ϭ 15.4, 10.5, 4.5, 3.3 Hz, 1 H), 2.42 (dd, J ϭ 14.0, 9.8 Hz, 1 H),
warm to room temp. The organic phase was collected and the aque-
ous phase was extracted with EtOAc (3 ϫ 100 mL). The combined
organic layers were dried with MgSO4 and the solvents evaporated
under reduced pressure. Flash chromatography of the residual oil
with EtOAc/cyclohexane (1:9) gave complete separation of the two
diastereomers 4a (4.00 g, 56%) and 4b (1.62 g, 23%).
2.59 (ddd, J ϭ 14.0, 10.9, 3.2 Hz, 1 H), 2.90 (ddd, J ϭ 15.2, 10.9,
and 4.5 Hz, 1 H), 3.11 (ddd, J ϭ 15.3, 10.6, 5.2 Hz, 1 H), 3.38 (s,
3 H), 3.96 (dd, J ϭ 13.0, 10.3 Hz, 1 H), 4.21 (dd, J ϭ 9.8, 7.3 Hz,
1 H), 4.31 (dd, J ϭ 13.0, 4.7 Hz, 1 H), 4.58 (dd, J ϭ 10.2, 4.6 Hz,
1 H), 4.75 (d, J ϭ 7.2 Hz, 1 H), 4.96 (d, J ϭ 7.2 Hz, 1 H),
7.21Ϫ7.48 (m, 5 H) ppm. 13C NMR (CDCl3): δ ϭ 30.6 (CH2), 41.2
(CH2), 41.4 (CH), 48.4 (CH2), 55.8 (CH3), 62.5 (CH), 64.9 (CH2),
90.0 (CH2), 99.5 (Cq), 119.4 (Cq), 126.8 (CH), 127.5 (CH), 128.6
(CH), 138.4 (Cq) ppm.
Major Isomer (αS)-4a: Colourless oil. Rf ϭ 0.61 (EtOAc/cyclohex-
ane, 3:7). [α]2D4 ϭ Ϫ214 (c ϭ 1.0, CH2Cl2). MS: m/z ϭ 289 [MH]ϩ.
IR (film): ν˜ ϭ 2255 cmϪ1 1H NMR (CDCl3): δ ϭ 2.51 (d, J ϭ
.
8.0 Hz, 2 H), 3.30 (s, 3 H), 3.67 (t, J ϭ 8.2 Hz, 1 H), 3.96 (t, J ϭ Oxopiperidine Derivative 6: Pale yellow oil. Rf ϭ 0.24 (EtOAc/
8.0 Hz, 1 H), 4.02 (dd, J ϭ 7.1, 8.3 Hz, 1 H), 4.14 (d, J ϭ 2.3 Hz, cyclohexane, 3:7). MS: m/z ϭ 289 [MH]ϩ. IR (film): ν˜ ϭ 2348,
1
1 H), 4.28 (dd, J ϭ 7.1, 7.9 Hz, 1 H), 4.31 (d, J ϭ 2.4 Hz, 1 H), 1723 cmϪ1. H NMR (CDCl3): δ ϭ 2.49 (m, 2 H), 2.67 (m, 2 H),
4.50 (d, J ϭ 2.5 Hz, 1 H), 4.79 (d, J ϭ 6.2 Hz, 1 H), 4.84 (d, J ϭ 2.99 (dt, J ϭ 11.8, 2.9 Hz, 1 H), 3.25 (s, 3 H), 3.69 (m, 1 H),
2.5 Hz, 1 H), 4.91 (d, J ϭ 6.2 Hz, 1 H), 7.27Ϫ7.40 (m, 5 H) ppm.
3.78Ϫ3.84 (m, 2 H), 3.89Ϫ3.98 (m, 2 H), 4.58 (s, 2 H), 7.30Ϫ7.48
13C NMR (CDCl3): δ ϭ 38.7 (CH2), 49.4 (CH), 56.3 (CH3), 65.4 (m, 5 H) ppm. 13C NMR (CDCl3): δ ϭ 40.6 (CH2), 43.3 (CH2),
(CH), 74.3 (CH2), 82.3 (CH2), 88.4 (CH2), 93.9 (CH2), 116.6 (Cq), 45.6 (CH2), 51.7 (CH), 55.4 (CH3), 67.0 (CH), 70.1 (CH2), 96.5
127.7 (CH), 128.4 (CH), 128.8 (CH), 137.2 (Cq), 154.5 (Cq) ppm.
(CH2), 115.6 (Cq), 127.8 (CH), 128.4 (CH), 129.0 (CH), 138.9 (Cq),
C16H20N2O3 (288.3): calcd. C 66.65, H 6.99, N 9.72; found C 66.71, 204.0 (Cq) ppm. C16H20N2O3 (288.3): calcd. C 66.65, H 6.99, N
H 6.81, N 9.71.
9.72; found C 66.43, H 6.81, N 9.74.
Minor Isomer (αR)-4a: Colourless oil. Rf ϭ 0.54 (EtOAc/cyclohex-
Cyclic Enaminone 7: A 0.1  aqueous hydrochloric acid solution
ane, 3:7). 1H NMR (CDCl3): δ ϭ 2.20 (dd, J ϭ 9.7, 14.0 Hz, 1 H), (5 mL) was added to a solution of 5a/5b mixture (240 mg,
2.35 (dd, J ϭ 6.0, 14.0 Hz, 1 H), 3.49 (s, 3 H), 3.67 (dd, J ϭ 6.5, 0.83 mmol) in dioxane (2 mL). After stirring at room temp. for 1 h,
8.3 Hz, 1 H), 3.89 (d, J ϭ 2.2 Hz, 1 H), 4.16 (dd, J ϭ 6.0, 9.7 Hz, the mixture was extracted with CH2Cl2 (3 ϫ 10 mL). The combined
1 H), 4.19 (d, J ϭ 2.3 Hz, 1 H), 4.24 (dd, J ϭ 6.8, 7.0 Hz, 1 H), organic layers were dried with MgSO4 and the solvents evaporated
4.40 (m, 1 H), 4.56 (d, J ϭ 4.4 Hz, 1 H), 4.74 (d, J ϭ 4.4 Hz, 1 H), under reduced pressure. Flash chromatography of the residual oil
4.90 (s, 1 H), 4.92 (s, 1 H), 7.11Ϫ7.49 (m, 5 H) ppm. 13C NMR
with EtOAc/MeOH (9:1) gave the single product 7 (150 mg, 83%)
(CDCl3): δ ϭ 39.1 (CH2), 52.7 (CH), 56.2 (CH3), 63.1 (CH), 75.0 as a yellow oil which degraded rapidly with time. Rf ϭ 0.34 (EtOAc/
(CH2), 86.9 (CH2), 88.4 (CH2), 93.7 (CH2), 117.8 (Cq), 126.8 (CH), MeOH, 9:1). MS: m/z ϭ 218 [MH]ϩ. IR (film): ν˜ ϭ 3366, 1575
1
127.7 (CH), 128.6 (CH), 141.2 (Cq), 154.1 (Cq) ppm.
cmϪ1. H NMR (CDCl3): δ ϭ 2.37 (m, 2 H), 3.21 (dt, J ϭ 13.2,
6.7 Hz, 1 H), 3.28 (m, 1 H), 3.86 (m, 2 H), 4.41 (dd, J ϭ 7.3,
5.6 Hz, 1 H), 4.85 (d, J ϭ 7.4 Hz, 1 H), 7.18 (d, J ϭ 7.4 Hz, 1 H),
7.21Ϫ7.39 (m, 5 H) ppm. 13C NMR (CDCl3): δ ϭ 35.4 (CH2), 44.9
(CH2), 69.6 (CH), 61.8 (CH2), 97.9 (CH), 127.2 (CH), 128.6 (CH),
129.1 (CH), 136.4 (Cq), 154.3 (CH), 192.2 (Cq) ppm.
Cyclisation Products 5 and 6: Under argon, BF3·Et2O (2.90 mL,
23.0 mmol) was added dropwise to a stirred solution of (αS)-4a
(4.42 g, 15.3 mmol) in diethyl ether (20 mL) at Ϫ70 °C. After stir-
ring at this temperature for 1 h, a saturated aqueous solution of
NaHCO3 (40 mL) was added and the mixture allowed to warm to
room temp. The organic phase was collected and the aqueous phase
was extracted with EtOAc (3 ϫ 50 mL). The combined organic
Amide Diastereoisomers 8: A solution of a 5a/5b mixture (1.46 g,
4.76 mmol) in methanol was treated with solid K2CO3 (0.37 g,
Eur. J. Org. Chem. 2003, 268Ϫ273
271

K. Partogyan-Halim, L. Besson, D. J. Aitken, H.-P. Husson
2.67 mmol) followed by 30% hydrogen peroxide solution (2.4 mL, was added. The mixture was stirred vigorously under hydrogen (1
FULL PAPER
19.0 mmol). After stirring at room temp. for 10 min, the mixture
was heated at 40 °C for 24 h. The mixture was then concentrated
atm) for 4 h. The catalyst was removed by filtration through a plug
of Celite, which was washed copiously with water. The filtrate was
by evaporation of methanol under reduced pressure, then diluted concentrated to dryness under reduced pressure and the pale yellow
with brine (10 mL). The resulting aqueous phase was extracted
with CH2Cl2 (5 ϫ 10 mL). The combined organic layers were dried
with MgSO4 and the solvents evaporated under reduced pressure
to give a mixture of epimers of 8 as a viscous oil (1.25 g, 86%),
sufficiently pure for use directly in the next step. Attempted flash
chromatography with EtOAc/cyclohexane (7:3) was inefficient and
wasteful, since one epimer could not be eluted from the column.
solid residue was dissolved in a minimum of water and the solution
applied to a column of Dowex 50X8Ϫ100 ion exchange resin (Hϩ
form). The column was eluted with water until the eluent was neut-
ral; subsequent elution with 1  NH4OH solution and concentra-
tion of the eluent gave the title compound as a white amorphous
solid, isolated as its hydrate form 10 (105 mg, 70%). [α]2D4 ϭ Ϫ17
(c ϭ 1.0, H2O). MS: m/z ϭ 162 [MH]ϩ. IR (nujol): ν˜ ϭ 3415, 1625
MS: m/z ϭ 307 [MH]ϩ. IR (CH2Cl2): ν˜ ϭ 3497, 1682 cmϪ1. Mobile cmϪ1
.
1H NMR (D2O): δ ϭ 1.59Ϫ1.78 (m, 3 H), 2.10 (m, 1 H),
2.89 (dd, J ϭ 12.9, 4.0 Hz, 1 H), 3.17 (dd, J ϭ 12.9, 4.0 Hz, 1 H),
Isomer: Rf ϭ 0.35 (EtOAc/cyclohexane, 7:3). 1H NMR (CDCl3):
δ ϭ 1.89 (ddd, J ϭ 13.7, 10.0, 4.0 Hz, 1 H), 2.31 (m, 1 H), 2.45 (d, 3.49 (m, 1 H) ppm. 13C NMR (CDCl3): δ ϭ 36.0 (CH2), 39.3
J ϭ 8.3 Hz, 2 H), 3.12 (m, 1 H), 3.30 (ddd, J ϭ 15.4, 10.1, 5.8 Hz, (CH2), 43.6 (CH2), 57.3 (CH), 93.3 (Cq), 173.3 (Cq) ppm.
1 H), 3.40 (s, 3 H), 3.58 (t, J ϭ 8.4 Hz, 1 H), 3.97Ϫ4.13 (m, 2 H),
4.28 (dd, J ϭ 12.0, 3.9 Hz, 1 H), 4.80 (d, J ϭ 7.1 Hz, 1 H), 4.91
(d, J ϭ 7.1 Hz, 1 H), 6.96 (d, J ϭ 4.1 Hz, 1 H), 7.08 (d, J ϭ 4.3 Hz,
Acknowledgments
1 H), 7.18Ϫ7.36 (m, 5 H) ppm. 13C NMR (CDCl3): δ ϭ 32.8 (CH2),
This work was supported in part by a grant to K. P.-H. from the
34.3 (CH2), 45.1 (CH2), 52.9 (CH), 55.3 (CH3), 65.7 (CH2), 67.4
´
Chancellerie des Universites de Paris. We thank Drs. J.-C. Florent
(CH), 89.6 (CH2), 100.8 (Cq), 126.7 (CH), 127.2 (CH), 128.4 (CH),
139.2 (Cq), 175.8 (Cq) ppm. Non-eluting Isomer: 1H NMR
(CDCl3): δ ϭ 1.26 (m, 2 H), 2.32 (d, J ϭ 9.3 Hz, 2 H), 2.80 (m, 1
H), 3.11 (m, 1 H), 3.37 (s, 3 H), 3.70 (m, 2 H), 4.09 (m, 2 H), 4.55
(d, J ϭ 6.9 Hz, 1 H), 4.98 (d, J ϭ 6.9 Hz, 1 H), 6.48 (br. s, 1 H),
6.57 (br. s, 1 H), 7.20Ϫ7.39 (m, 5 H) ppm. 13C NMR (CDCl3): δ ϭ
30.8 (CH2), 38.0 (CH2), 43.5 (CH2), 55.7 (CH3), 59.7 (CH), 62.6
(CH2), 65.7 (CH), 90.0 (CH2), 100.7 (Cq), 126.2 (CH), 127.3 (CH),
128.7 (CH), 138.7 (Cq), 172.8 (Cq) ppm.
and J. Royer for useful discussions on synthetic acetone equivalents,
and F. Libot for recording the mass spectra. D. J. A. is the benefi-
ciary of CNRS research funding (AIP).
[1]
H. Kessler, M. Kühn, T. Löschner, Liebigs Ann. Chem. 1986,
1Ϫ20.
[2]
H. S. Andersen, T. K. Hansen, J. Lau, N. P. H. Møller, O. H.
Olsen, F. Axe, F. Bakir, Y. Ge, D. D. Holsworth, L. M. Judge,
M. J. Newman, R. T. Uyeda, B. Z. Shapira, Preparation of 2-
oxalylaminothieno[2,3-c]pyridines as modulators of protein
tyrosine phosphatases (ptpases), PCT Int. Appl. WO 0204459,
2002; Chem. Abstr. 2002, 136, 118439.
Lactone 9: A 1  NaOH solution (10 mL) was added to a solution
of amide 8 as a mixture of epimers (1.11 g, 3.61 mmol) in MeOH
(20 mL) and the mixture was stirred at room temp. for 18 h. The
mixture was then concentrated by evaporation of methanol under
reduced pressure, and 6  HCl was added dropwise until pH ϭ 1
was reached. MeOH (2 mL) was then added and the mixture was
stirred at room temp. for 18 h. After this time, the mixture was
extracted with CH2Cl2 (3 ϫ 20 mL). The combined organic layers
were dried with MgSO4 and the solvents evaporated under reduced
pressure. Flash chromatography of the residual oil with EtOAc/
cyclohexane (3:7) gave the lactone 9 which was recrystallised from
diisopropyl ether (0.64 g, 72%). M.p. 141 °C (diisopropyl ether).
Rf ϭ 0.25 (EtOAc/cyclohexane, 3:7). [α]2D4 ϭ Ϫ148 (c ϭ 1.0,
[3] [3a]
P. L. Ornstein, M. B. Arnold, W. H. W. Lunn, L. J. Heinz,
J. D. Leander, D. Lodge, D. D. Schoepp, Bioorg. Med. Chem.
Lett. 1998, 8, 389Ϫ394.
[3b]
J. W. Skiles, P. P. Giannousis, K.
R. Fales, Bioorg. Med. Chem. Lett. 1996, 6, 963Ϫ966.
D. Gustafsson, J.-E. Nyström, Preparation of amino acid deriv-
atives and their use as thrombin inhibitors, PTC Int. Appl. WO
9746577, 1997; Chem. Abstr. 1998, 128, 75673.
[4]
[5] [5a]
C. A. Brooks, D. L. Comins, Tetrahedron Lett. 2000, 41,
[5b]
3551Ϫ3553.
F. P. J. T. Rutjes, J. J. N. Veerman, W. J. N.
Meester, H. Hiemstra, H. E. Schoemaker, Eur. J. Org. Chem.
[5c]
ˆ
M. Haddad, M. Larcheveque, Tetrahed-
1999, 1127Ϫ1135.
[5d]
ron: Asymmetry 1999, 10, 4231Ϫ4237.
ham, P. C. Anderson, P. L. Beaulieu, T. Bogri, Y. Bousquet, L.
J. Gillard, A. Abra-
CH2Cl2). MS: m/z ϭ 246 [MH]ϩ. IR (film): ν˜ ϭ 1725, 1713 cmϪ1
.
1H NMR (CDCl3): δ ϭ 2.12 (dd, J ϭ 12.3, 2.6 Hz, 1 H), 2.30 (d,
J ϭ 15.2 Hz, 1 H), 2.51 (m, 1 H), 2.74 (dd, J ϭ 15.3, 12.4 Hz, 1
H), 3.08 (m, 2 H), 3.37 (dd, J ϭ 12.2, 3.0 Hz, 1 H), 3.76 (dd, J ϭ
10.7, 3.5 Hz, 1 H), 4.30 (dd, J ϭ 11.2, 3.5 Hz, 1 H), 4.43 (dd, J ϭ
10.7, 11.2 Hz, 1 H), 7.30Ϫ7.41 (m, 5 H) ppm. 13C NMR (CDCl3):
δ ϭ 40.7 (CH2), 43.0 (CH2), 50.5 (CH2), 62.3 (CH), 63.8 (CH2),
72.8 (CH), 128.1 (CH), 129.2 (CH), 129.3 (CH), 135.4 (Cq), 167.6
(Cq), 205.7 (Cq) ppm. C14H15N2O3 (245.3): calcd. C 68.56, H 6.16,
N 5.71; found C 68.44, H 6.16, N 5.97. In one run, traces of a
second component were isolated during the chromatography, for
Grenier, I. Guse, P. Lavalee, J. Org. Chem. 1996, 61,
´
2226Ϫ2231.
[6] [6a]
A. S. Golubev, H. Schedel, G. Radics, J. Seiler, K. Burger,
[6b]
Tetrahedron Lett. 2001, 42, 7941Ϫ7944.
S. Carbonnel, C.
Fayet, J. Gelas, Y. Troin, Tetrahedron Lett. 2000, 41,
[6c]
8293Ϫ8296.
G. Galley, B. Zeimer, M. Pätzel, J. Prakt.
[6d]
Chem. 1998, 340, 551Ϫ556.
J. Barluenga, F. Aznar, C.
[6e]
´
Valdes, C. Ribas, J. Org. Chem. 1998, 63, 3918Ϫ3924.
G.
R. Heintzelman, S. M. Weinreb, M. Parvez, J. Org. Chem.
1996, 61, 4594Ϫ4599.
[7] [7a]
C. Mühlemann, P. Hartmann, J.-P. Obrecht, Org. Synth.
[7b]
1
1993, 71, 200Ϫ206.
(Weinheim) 1992, 325, 419Ϫ423.
Schoepp, M. B. Arnold, J. D. Leander, D. Lodge, J. W. Paschal,
C. Herdeis, W. Engel, Arch. Pharm.
which the structure epi-9 was proposed on the basis of H NMR
[7c]
spectroscopic data. 1H NMR (CDCl3): δ ϭ 2.31 (dt, J ϭ 15.1,
4.1 Hz, 1 H), 2.49 (ddd, J ϭ 15.1, 10.2, 5.5 Hz, 1 H), 2.68 (ddd,
J ϭ 12.7, 10.2, 4.0 Hz, 1 H), 2.83 (d, J ϭ 7.9 Hz, 2 H), 3.18 (ddd,
J ϭ 12.7, 5.5, 4.5 Hz, 1 H), 3.86 (t, J ϭ 7.9 Hz, 1 H), 4.19 (dd, J ϭ
7.0, 4.3 Hz, 1 H), 4.50 (dd, J ϭ 11.3, 7.0 Hz, 1 H), 4.64 (dd, J ϭ
11.3, 4.3 Hz, 1 H), 7.38 (m, 5 H) ppm.
P. L. Ornstein, D. D.
[7d]
T. Elzey, J. Med. Chem. 1991, 34, 90Ϫ97.
A. A. Molerino,
D. G. I. Kingston, J. W. Reed, J. Nat. Prod. 1989, 52, 99Ϫ108.
[7e]
P. Hartmann, J.-P. Obrecht, Synth. Commun. 1988, 18,
553Ϫ557.
[8]
[9]
`
G. Jolles, G. Poignet, J. Robert, B. Terlain, J.-P. Thomas, Bull.
Soc. Chim. Fr. 1965, 2252Ϫ2259.
F. Machetti, F. M. Cordero, F. De Sarlo, A. Brandi, Tetrahed-
ron 2000, 57, 4995Ϫ4998.
(S)-4-Oxopipecolic Acid Hydrate (10): Lactone
9 (230 mg,
0.93 mmol) was dissolved in 1  HCl (2 mL) and 10% Pd/C (90 mg)
272
Eur. J. Org. Chem. 2003, 268Ϫ273

Asymmetric Synthesis of (S)-4-Oxopipecolic Acid by a 3ϩ3 Atom-Unit Assembly Strategy
FULL PAPER
[10]
´
F. Machetti, F. M. Cordero, F. De Sarlo, A. Guarna, A.
M. Le Bail, J. Perard, D. J. Aitken, M. Bonin, H.-P. Husson,
Tetrahedron Lett. 1997, 38, 7177Ϫ7180.
Brandi, Tetrahedron Lett. 1996, 37, 4205Ϫ4208.
[11] [11a]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
A. Golubev, N. Sewald, K. Burger, Tetrahedron 1996, 52,
D. J. Aitken, L. Besson, K. Partogyan-Halim, H.-P. Husson,
Heterocycl. Commun. 1999, 5, 529Ϫ532.
F. Vergne, D. J. Aitken, H.-P. Husson, J. Chem. Soc., Perkin
Trans. 1 1991, 1346Ϫ1347.
C. Yue, J. Royer, H.-P. Husson, J. Org. Chem. 1992, 57,
[11b]
14757Ϫ14776.
A. Golubev, N. Sewald, K. Burger, Tetra-
hedron Lett. 1995, 36, 2037Ϫ2040.
Y. Bousquet, P. C. Anderson, T. Bogri, J.-S. Duceppe, L. Gren-
[12]
[13]
[14]
[15]
ier, I. Guse, Tetrahedron 1997, 53, 15671Ϫ15678.
´
´
R. Badorrey, C. Cativiela, M. D. Dıaz-da-Villegas, J. A. Gal-
vez, Tetrahedron Lett. 1997, 38, 2547Ϫ2550.
R. F. W. Jackson, L. J. Graham, A. B. Rettie, Tetrahedron Lett.
1994, 35, 4417Ϫ4418.
4211Ϫ4214.
X.-P. Gu, I. Ikeda, M. Okahara, J. Org. Chem. 1987, 52,
3192Ϫ3196.
J. L. Marco, J. Royer, H.-P. Husson, Tetrahedron Lett. 1985,
F. Couty, Amino Acids 1999, 16, 297Ϫ320.
26, 3567Ϫ3570.
[16] [16a]
[16b]
S. Laschat, T. Dickner, Synthesis 2000, 1781Ϫ1813.
P.
M. Cellier, Y. Gelas-Mialhe, H.-P. Husson, B. Perrin, R. Remu-
son, Tetrahedron: Asymmetry 2000, 11, 3913Ϫ3919.
L. Besson, M. Le Bail, D. J. Aitken, H.-P. Husson, F. Rose-
Munch, E. Rose, Tetrahedron Lett. 1996, 37, 3307Ϫ3308.
D. Bailey, P. A. Millwood, P. D. Smith, Chem. Commun. 1998,
[16c]
633Ϫ640.
M. Rubiralta, E. Giralt, A. Diez, Piperidine:
Structure, Preparation, Reactivity and Synthetic Applications of
Piperidine and its Derivatives, Elsevier, Amsterdam, 1991.
H.-P. Husson, J. Royer, Chem. Soc. Rev. 1999, 28, 383Ϫ394,
and references therein.
´
M. Le Bail, J. Perard, D. J. Aitken, H.-P. Husson, Tetrahedron
[17]
[18]
Lett. 1999, 40, 5309Ϫ5313.
Received August 6, 2002
[O02460]
For leading references on previous work with compound 2, see:
Eur. J. Org. Chem. 2003, 268Ϫ273
273