Enantioselective approach to 3-substituted prolines

Article abstract of DOI:10.1016/S0040-4039(01)01869-X

Enantioselective synthesis of 3-substituted prolines was achieved starting from commercially available 3-hydroxy-(S)-2-proline. Palladium-mediated couplings were used to introduce a variety of groups at C3 using the corresponding enol triflate derived from N-trityl 3-oxo-(S)-2-proline methyl ester. Cleavage of the trityl residue and hydrogenation provided final products with good to modest diastereoselectivity.

Full text of DOI:10.1016/S0040-4039(01)01869-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 8571–8573
Enantioselective approach to 3-substituted prolines
Theodore M. Kamenecka,* You-Jung Park, Linus S. Lin, Thomas Lanza, Jr. and
William K. Hagmann
Department of Medicinal Chemistry, Merck Research Laboratories, PO Box 2000, Rahway, NJ 07065, USA
Received 17 August 2001; accepted 25 September 2001
Abstract—Enantioselective synthesis of 3-substituted prolines was achieved starting from commercially available 3-hydroxy-(S)-2-
proline. Palladium-mediated couplings were used to introduce a variety of groups at C3 using the corresponding enol triflate
derived from N-trityl 3-oxo-(S)-2-proline methyl ester. Cleavage of the trityl residue and hydrogenation provided final products
with good to modest diastereoselectivity. © 2001 Elsevier Science Ltd. All rights reserved.
Introduction of rigidity into bioactive peptides has been
a useful tool to study the conformational requirements
for biological activity.1 Proline is well known for its
ability to introduce conformational restrictions into
bioactive peptides by forming cis peptide bonds which
can induce the formation of b-turns as well as influence
protein folding.2,3 Replacement of proline with substi-
tuted analogs provides additional information about
receptor recognition and affinity. Despite the growing
interest in using substituted prolines as molecular
probes, there are few practical methods to prepare
them.
more, many of these routes provide racemic material
that needs to be resolved at a later stage. We desired an
approach which would provide rapid access to a variety
of analogs from a common intermediate in enantiomer-
ically pure form. Herein, we describe such a convergent
approach starting from commercially available 3-(R)-
hydroxy-2-(S)-proline (Scheme 1).
Esterification and trityl protection of 3-hydroxyproline
afforded hydroxyester 1 uneventfully. Oxidation using
catalytic TPAP provided ketone 2. This b-ketoester was
deprotonated with KHMDS and trapped as its enol
triflate to give the D 3-olefin.5 The regioselectivity of the
triflate formation is surprising given the lower pKa of
the proton at C2, however, trityl protection of a-amino
esters is known to shield the chiral center from deproto-
nation.6a–c This enol triflate smoothly participates in a
number of palladium-mediated couplings as illustrated
in Table 1.
Recently, we had the need for a number of 3-substi-
tuted proline derivatives in enantiomerically pure form.
Although there are several literature procedures for
their synthesis, none of them proved optimal.4a–f Most
approaches start with acyclic precursors and require
several steps for each individual substrate. Further-
OH
OH
O
i. HCl, MeOH
(nPr)4NRuO4
CO2H
CO2Me
CO2Me
N
H
N
Tr
90%
N
Tr
ii. TMSCl; TrCl;
K2CO3, MeOH
1
2
OTf
R
KHMDS
palladium
coupling
Cl
CO2Me
CO2Me
N
Tr
N
Tr
N
NTf2
3
92%
Scheme 1.
* Corresponding author. Tel.: (858) 452-5892 (x449); fax: (858) 454-3792; e-mail: theodore kamenecka@merck.com
–
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01869-X

8572
T. M. Kamenecka et al. / Tetrahedron Letters 42 (2001) 8571–8573
Table 1.
explanation for the good to excellent diastereoselectiv-
ity obtained by hydrogenating the amine salts comes
from A1,2 strain. As the C3 vinyl substituent gets larger,
A1,2 strain increases as does the preference for the C2
carboxylate to adopt an axial conformation. With the
ester blocking the b-face, hydrogenation comes from
below giving the 2,3-cis product.
R
OTf
CO2Me
palladium
coupling
CO2Me
N
Tr
N
Tr
3
4a-j
Reagent
Product
Yieldc (%)
To determine if any racemization had occurred during
its synthesis, amine salt 5b was converted to its t-butyl
carbamate (6b) using di-tert-butyl dicarbonate and tri-
ethylamine. For comparison, racemic material was pre-
pared following standard literature procedures.4f
Compound 6b was >98% enantiomerically pure as
judged by chiral HPLC analysis indicating that no
racemization occurred at the b-keto ester stage or dur-
ing enol triflate formation.13
4-MeO2CPh-Xa,b
Ph-X
4-Pyridyl-X
trans-Hexenyl-X
3-Thiophenyl-X
p-Tolyl-Y
Vinyl-Y
Me-Y
Me3Sn-Y
CO, MeOH
4-MeO2CPh (4a)
Ph (4b)
84
74
44
88
57
70
86
65
54
67
4-Pyridyl (4c)
1-E-Hexenyl (4d)
3-Thiophenyl (4e)
p-Tolyl (4f)
Vinyl (4g)
Me (4h)
SnMe3 (4i)
CO2Me (4j)
In summary, we have developed a convergent and
efficient procedure for the synthesis of 3-substituted
prolines in enantiomerically pure form. It has the
advantage of proceeding through a common intermedi-
ate (enol triflate 3) which is readily available in two
steps from commercially available material. Palladium-
mediated couplings proceed in good yield and modest
to high diastereoselectivity can be achieved in the
hydrogenation step. These methods should allow rapid
access to analogs which were previously prepared by
more laborious routes.
a X=B(OH)2, Y=SnMe3.
b Proline benzyl ester was used in this case.
c Yields are unoptimized.
A variety of palladium mediated couplings were exam-
ined and gave good to modest yields of coupled prod-
ucts.7,8 Suzuki couplings of aryl and vinyl boronic acids
proceeded smoothly as did a number of Stille couplings
of both aryl and alkyl stannanes.9,10 Conversion of the
enol triflate to the vinyl stannane 4i was also feasible
and may thereby expand the utility of this substrate in
palladium mediated reactions. Lastly, standard car-
bonylation proved uneventful giving the diester 4j.
These examples attest to the versatility of 3 as a willing
partner in palladium-catalyzed couplings.
Acknowledgements
We wish to thank Mr. Henry Murillo and Dr. Nathan
Yates for LC-MS measurements.
Catalytic hydrogenation of the coupled products satu-
rated the olefin and cleaved the trityl protecting group
giving 3-substituted proline methyl esters in good yield
(Table 2). Unexpectedly, varying degrees of diastereo-
selectivity were observed with several substrates with no
clear trend.11 To find a more general procedure, the
trityl group was removed with HCl(g) in MeOH and
the crude amine salts were subjected to hydrogenation.
This procedure consistently provided the 2,3-cis
product as the major diastereomer.12 One possible
References
1. (a) Momany, F. A.; Chuman, H. Methods Enzymol. 1986,
124, 3; (b) Marshall, G. R. In Chemical Recognition in
Biological Systems; Creighton, A. M.; Turner, S., Eds.;
The Chemical Society: London, 1982; p. 278.
2. (a) Baures, P. W.; Ojala, W. J.; Gleason, W. B.; Johnson,
R. L. J. Pept. Res. 1997, 50, 1; (b) Halab, L.; Lubell, W.
D. J. Org. Chem. 1999, 64, 3312.
Table 2.
R
i. HCl, MeOH
CH2Cl2;
R
3. Beausoleil, E.; Sharma, R.; Michnick, S. W.; Lubell, W.
D. J. Org. Chem. 1998, 63, 6572.
+
CO2Me
CO2Me
N
Tr
N
4. (a) Blanco, M.-J.; Paleo, M. R.; Penide, C.; Sardina, F. J.
J. Org. Chem. 1999, 64, 8786; (b) Sasaki, N. A.; Dockner,
M.; Chiaroni, A.; Riche, C.; Potier, P. J. Org. Chem.
1997, 62, 765; (c) Koskinen, A. M. P.; Schwerdtfeger, J.;
Edmonds, M. Tetrahedron Lett. 1997, 38, 5399; (d)
Sharma, R.; Lubell, W. D. J. Org. Chem. 1996, 61, 202;
(e) Waid, P. P.; Flynn, G. A.; Huber, E. W.; Sabol, J. S.
Tetrahedron Lett. 1996, 37, 4091; (f) Chung, J. Y. L.;
Wasicak, J. T.; Arnold, W. A.; May, C. S.; Nadzan, A.
M.; Holladay, M. W. J. Org. Chem. 1990, 55, 270.
5. The current method described is analogous to the strat-
egy used by others for the preparation of 3- and 4-substi-
tuted prolines starting from 4-oxoproline.4a,c,d
Cl-
ii. H2, Pd/C,
MeOH
H2
5
4
Structure
R
Product
cis/transa
4b
4d
4e
4h
4j
Ph
5b
5db
5e
5h
5j
\15:1
4:1
13:1
5:1
2-Hexenyl
3-Thiophenyl
Me
CO2Me
2:5:1
a As determined by 1H NMR.
b R=n-hexyl.

T. M. Kamenecka et al. / Tetrahedron Letters 42 (2001) 8571–8573
8573
6. (a) Baldwin, J. E.; North, M.; Flinn, A.; Moloney, M. G.
Tetrahedron 1989, 45, 1453; (b) Rapoport, H.; Lubell, W.
D. J. Am. Chem. Soc. 1988, 110, 7447; (c) Garst, M. E.;
Bonfiglio, J. N.; Grudoski, D. A.; Marks, J. J. Org.
Chem. 1980, 45, 2307.
3, 10 mol% Pd(OAc)2, 20 mol% dppf, 2 equiv. TEA, 30
equiv. MeOH in DMF at 60°C for 18 h under a balloon
of CO pressure.
9. (a) Stille, J. K. Angew. Chem. 1986, 98, 504; (b) Stille, J.
K. Pure Appl. Chem. 1985, 57, 1771.
7. Representative procedure: To a solution of 1.0 g (1.93
mmol) of 3 and 0.60 g (3.9 mmol) of phenyl boronic acid
in toluene:MeOH (10:1, v:v, 11 mL) was added 0.40 g
(2.9 mmol) of K2CO3 and 79 mg (5 mol%) of
PdCl2(dppf). The reaction mixture was stirred at 85°C for
9 h, cooled and concentrated. Purification by silica-gel
chromatography (3:1 hexanes:Et2O) provided 0.62 g
(72%) of 4b as a colorless solid, homogeneous by TLC
analysis.
8. (a) For the Stille couplings, the following conditions were
used: 1 equiv. 3, 2 equiv. Stannane, using (1) 10 mol%
Pd2dba3, 20 mol% Ph3As, in NMP at 50°C, or (2) 4
equiv. LiCl, 10 mol% Pd(Ph3P)4 in THF at 80°C; (b)
conditions for the carbonylation were as follows: 1 equiv.
10. Suzuki, A. J. Organomet. Chem. 1999, 576, 147.
11. Direct hydrogenation of trityl protected derivatives 4a–
4h, 4j gave cis/trans mixtures of deprotected products
with the 2,3-cis predominating sometimes and other
times, the 2,3-trans product was major.
12. As determined by 1H NMR analysis. In the case of 5b
and 5h, comparison to authentic material confirmed the
assignments.
13. Chiral HPLC conditions: Chiralcel OJ 4.6×250 mm, 10
micron, 2% EtOH in hexane, isocratic at 0.75 mL/min.
Racemic material appears as two peaks: peak 1 (retention
time=15.5 min), peak 2 (retention time=23.6 min).
Analysis of 6b gave a single peak (retention time=23.8
min).