Synthesis of (2S,3R)- and (2S,3S)-[3-2H1]-proline via highly stereoselective hydrolysis of a silyl enol ether

Article abstract of DOI:10.1016/j.tetlet.2005.04.126

A straightforward synthesis of (2S)-[3,3-²H₂]-proline 1c and (2S,3R)- and (2S,3S)-[3-²H₁]-proline, 1b and 1a, respectively, has been devised. The key step of the route to the latter compounds involves highly stereoselective hydrolysis of the silyl enol ethers 3 and 3a, respectively, with protonation (deuteriation) from the re-face of the silyl enol ether.

Full text of DOI:10.1016/j.tetlet.2005.04.126

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 4653–4655
Synthesis of (2S,3R)- and (2S,3S)-[3-2H1]-proline via
highly stereoselective hydrolysis of a silyl enol ether
Paul Barraclough, Caroline A. Spray and Douglas W. Young*
Department of Chemistry, University of Sussex, Falmer, Brighton, BN1 9QJ, UK
Received 16 February 2005; accepted 27 April 2005
Available online 23 May 2005
Abstract—A straightforward synthesis of (2S)-[3,3-2H2]-proline 1c and (2S,3R)- and (2S,3S)-[3-2H1]-proline, 1b and 1a, respectively,
has been devised. The key step of the route to the latter compounds involves highly stereoselective hydrolysis of the silyl enol ethers 3
and 3a, respectively, with protonation (deuteriation) from the re-face of the silyl enol ether.
Ó 2005 Elsevier Ltd. All rights reserved.
The use of stereospecifically labelled amino acids in meta-
bolic studies1 and, in combination with multidimen-
sional NMR spectroscopy, in obtaining detailed
protein solution structures2–4 makes the ready availabi-
lity of such compounds of considerable importance.
The amino acid proline 1 and its post-translationally
modified derivatives are constituents of antibiotics and
are important for conformational constraint in proteins.
Several syntheses of samples of this amino acid, stereo-
specifically labelled at C-3 have been completed1,5,6
including a chemicoenzymatic synthesis by ourselves.6
We now wish to report a straightforward chemical syn-
thesis of samples of proline labelled with deuterium in
the 3-pro-R or in the 3-pro-S position
hydrolysis of a solution of the silyl enol ether 3 in THF
containing excess trimethylsilyl chloride by adding
2
excess H2O gives remarkable stereoselectivity in deute-
riation at C-3 as shown in Scheme 1 below. Although
the 1H NMR spectrum of the unpurified product 2a
was complicated by the well-known8 conformational
isomerism shown by N-acylprolines, only one of the
two signals at 2.50 and 2.89 ppm due to the diastereo-
topic protons at C-3 in the unlabelled parent 2 was pres-
ent. Since the signal at 2.89 ppm in the unlabelled
compound showed a 4.5% enhancement on irradiation
of the two doublets due to H-2 at ca. 4.6 ppm it could
be assigned to H-3R and so the signal remaining at
2.5 ppm in the deuteriated compound 2a was due to
the proton, H-3S. Deuteriation had therefore occurred
from the re-face of the silyl enol ether 3.
4
3
2
5
Attempts to purify compound 2a by chromatography
using silica gel Ôwashed outÕ the label and so the unpuri-
fied compounds 2 and 2a were reduced directly to alco-
hols 4 and 4a using sodium borohydride in methanol
and diethyl ether at 0 °C, as shown in Scheme 1. This
allowed the stereoselectivity to be confirmed using a
CO2H
N
H
1
During work on the synthesis of analogues of kainic
acid,7 we prepared the silyl enol ether 3 from the 4-
ketoproline derivative 2 with a view of adding a side
chain at C-3 specifically from the re-face. We were
unsuccessful in this endeavour but have now found that
1
purer and more stable compound. The H NMR spec-
trum of the purified unlabelled compound 4 in C2HCl3
was complicated by conformational isomerism but a sim-
pler spectrum was obtained for a solution in C2H3CN at
60 °C. NOE experiments in C2HCl3, summarised in
Figure 1, allowed assignment of the peaks due to hydro-
gens, H-3R and H-3S, since irradiation at 4.2 ppm (H-2)
Keywords: Proline; Stereospecific deuteriation; Silyl enol ether;
Metabolism.
*
Corresponding author. Tel.: +44 01273 678327; fax: +44 01273
677196; e-mail: d.w.young@sussex.ac.uk
This compound had the required analytical and spectroscopic
properties.
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.04.126

4654
P. Barraclough et al. / Tetrahedron Letters 46 (2005) 4653–4655
H
HS
HR
H
O
O
HO
TMSO
2H
(i)
CO2t-Bu
(ii)
(iii)
t-Bu
t-Bu
2H
CO2
CO2
t-Bu
CO2
N
Boc
N
N
N
Boc
Boc
Boc
2
3
2a
4
4a, HR
=
Scheme 1. Reagents and conditions: (i) (a) LDA/THF/1 h, À78 °C, (b) Me3SiCl/À78 °C then 1 h, rt; (ii) 2H2O rt (98% from 2); (iii) NaBH4/MeOH/
Et2O/0 °C, 5 min then 30 min, rt (98% 4 directly from 2, 88% 4a from 2a as shown).
9.2%
1.6%
11.1%
Reasoning that the isotopomer 2b might be obtained by
protonation of the deuteriated silyl enol ether 3a, we set
about the preparation of the [3,3-2H2]-ketone 2c. We
therefore treated the ketone 2 with LDA in THF and
quenched the solution with 2H2O. Although the 1H
NMR spectrum of the product showed that any label
introduced at C-3 had been Ôwashed outÕ on purification
using silica gel chromatography, considerable deuteria-
tion had occurred at C-5. Since the label at C-5 had
remained after chromatography, we decided to Ôwash
inÕ deuterium at C-3 under conditions similar to those
that had ÔwashedÕ it out. The ketone 2 was therefore stir-
3.3%
HO
HS
H
HR
H
CO2t-Bu
N
Boc
Figure 1. NOE experiments on compound 4.
gave a considerably larger enhancement to the signal for
H-3 at 2.3 ppm than that at 2.0 ppm. This defined the
former signal as being due to H-3R. Irradiation at
4.3 ppm (H-4) gave a large enhancement to the signal
for H-3R, thus confirming the stereochemistry at C-4.
The 1H NMR spectrum of the labelled alcohol 4a exhibi-
ted a broad one-proton singlet for H-3S at 2.03 ppm and
a residual multiplet (ca. 20%) for H-3R at 2.3 ppm. The
specificity of the hydrolysis was thus confirmed. No deu-
terium was present in H-3S and the small amount of pro-
tium at H-3R presumably represented an isotope effect.
2
red in H2O containing silica gel which had previously
2
been washed with H2O as shown in Scheme 3. After
10 days at room temperature, almost complete exchange
was obtained and the product was converted to a solu-
tion of the silyl enol ether 3a using LDA followed by
reaction with excess trimethylsilyl chloride. Addition
of H2O to the solution gave the ketone 2b. The 1H
NMR spectrum of this compound exhibited H-3R at
2.89 ppm and a small signal for H-3S at 2.5 ppm.
Because of the acidity of the ketone it was not purified
but converted directly to the alcohol 4b by reduction
The high stereoselectivity found in this reaction is some-
what surprising but it is tempting to suggest that the
mechanism shown in Scheme 2 below could be partly
synchronous giving the chair-like transition state 7b
shown. Interaction of the bulky trimethylsilyl and tert-
butoxycarbonyl groups in 7b would result in proton-
ation occurring from the re-face as shown.
1
with NaBH4 as before. The H NMR spectrum of the
purified alcohol 4b showed a one-proton multiplet at
2.28 ppm for H-3R and ca. 30% residual protium at
2.0 ppm for H-3S. The residual protium presumably
reflected incomplete deuteriation in the initial dideuteri-
ated compound.
Me Me
Me Me
Me
Me Me
Me
Me
δ+
O
Me
Me
t
Me
O
H
O
O
O
O
O
BuO2 C
O
O
O
H
H
H
H
H
H
H
H
Si
H
Me
Si
O
O
Me
O
O
Si
O
O
Me
H
H
H
N
N
Boc
N
N
Me
Me
δ+
Me
H
Me
2H
2H
Me
Me
O
O
2H
2H
2H
H
H
H
2H
Me Me
Me Me
H
Me Me
O
2H
2H
3
7b
2a
7a
Scheme 2.
2H
2H
2H
HS
HR
2H
O
O
O
HO
TMSO
H
(iv)
(i)
(iii)
(ii)
CO2t-Bu
CO2t-Bu
CO2t-Bu
CO2t-Bu
N
N
CO2t-Bu
N
N
N
Boc
Boc
Boc
Boc
Boc
2
2c
3a
2b
4b
Scheme 3. Reagents and conditions: (i) 2H2O-silica gel/10 days/rt (93%); (ii) (a) LDA/THF À78 °C, 1 h, (b) Me3SiCl À78 °C, 5 min then 1 h, rt; (iii)
H2O rt (98% from 2c); (iv) NaBH4/MeOH/Et2O/0 °C, 5 min then 30 min, rt (40% 4b from 2b as shown; 89% 4c directly from 2c).

P. Barraclough et al. / Tetrahedron Letters 46 (2005) 4653–4655
4655
HR
H
HS
HR
HR
HS
HS
HR
HO
TsO
HS
CO2H
(ii)
(iii)
(i)
CO2t-Bu
CO2t-Bu
CO2t-Bu
N
N
N
N
Boc
Boc
H
Boc
4
1
5
6
2H
2H
1a HS
1b HR
1c HR = HS
=
=
2H
2H
5a HR
5b HS
5c HR = HS
=
=
2H
2H
6a HS
6b HR
6c HR = HS
=
=
2H
2H
4a HR
4b HS
4c HR = HS
=
=
=
2H
=
2H
=
2H
=
2H
Scheme 4. Reagents and conditions: (i) TsCl/pyridine/rt, 20 h (58% 5, 45% 5a, 71% 5b, 50% 5c); (ii) NaBH4/DMSO/85 °C, 6.5 h (94% 6, 87% 6a, 59%
6b, 89% 6c); (iii) 6 N HCl/rt, 2 h (98% 1, 30% 1a, 79% 1b, 57% 1c).
samples of proline, 1, 1a, 1b and 1c. The 1H NMR spec-
tra of these compounds are shown in Figure 2, (2S,3S)-
[3-2H1]-proline 1a having 22% protium at H-3S, and
(2S,3R)-[3-2H1]-proline having ca. 33% protium at
H-3R. The same synthetic route was used to convert
(2S)-[3,3-2H2]-4-ketoproline 2c to (2S)-[3,3-2H2]-proline
1c.
Acknowledgements
We thank the EPSRC for financial support and Profes-
sor J. R. Hanson for helpful discussions.
References and notes
1. Young, D. W. Top. Stereochem. 1994, 21, 381–465.
2. Kainosho, M. Nat. Struct. Biol. 1997, 4, 858–861.
3. Gardiner, K. H.; Kay, L. E. Biol. Magn. Reson. 1998, 16,
27–74.
4. Roberts, G. C. K. NMRof Macromolecules. A Practical
Approach; Oxford University Press: Oxford, 1993.
5. (a) Oba, M.; Terauchi, T.; Hashimoto, J.; Tanaka, T.;
Nishiyama, K. Tetrahedron Lett. 1997, 38, 5515–5518; (b)
Oba, M.; Terauchi, T.; Miyakawa, A.; Nishiyama, K.
Tetrahedron: Asymmetry 1999, 10, 937–945; (c) Oba, M.;
Miyakawa, A.; Nishiyama, K. J. Org. Chem. 1999, 64,
9275–9278.
6. Dieterich, P.; Young, D. W. Tetrahedron Lett. 1993, 34,
5455–5458.
7. Barraclough, P.; Hudhomme, P.; Spray, C. A.; Young, D.
W. Tetrahedron 1995, 51, 4195–4212.
8. Hondrelis, J.; Lonergan, G.; Voliotis, S.; Matsoukas, J.
Tetrahedron 1990, 46, 565–576.
Figure 2. 360 MHz 1H NMR spectra in C2H3O2H of (a) (2S)-proline
1; (b ) (2S,3S)-[3-2H1]-proline 1a; (c) (2S,3R)-[3-2H1]-proline 1b and (d)
(2S)-[3,3-2H2]-proline 1c.
Conversion of the unlabelled and labelled alcohols 4 to
the corresponding samples of proline 1 proceeded as
described in Scheme 4 above. The alcohols 4 were first
converted to the para-toluenesulfonates 5 using para-
toluenesulfonyl chloride in pyridine at room tempera-
ture. These were then reduced to the protected proline
derivatives 6
,à
by heating at 85 °C with NaBH4 in
DMSO for 6.5 h. Deprotection using 6 N aqueous
hydrochloric acid for 2 h at room temperature gave the
à Because of the change in group priority on removal of the oxygen
from C-4, the nomenclature requires that H-3S and H-3R be
redefined after step (ii) in Scheme 3.