Discovery of new peptide-based catalysts for the direct asymmetric aldol reaction

Article abstract of DOI:10.1016/S0960-894X(03)00498-0

A series of oligo-peptide based catalysts were prepared using Fmoc solid-phase peptide synthesis. It was found that peptides with N-terminal proline residues catalyzed an aldol reaction yielding enantiomeric enriched product. Peptide H-Pro-Glu-Leu-Phe-OH

Full text of DOI:10.1016/S0960-894X(03)00498-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 2445–2447
Discovery of New Peptide-Based Catalysts for the Direct
Asymmetric Aldol Reaction
Jacob Kofoed,a,b John Nielsenb and Jean-Louis Reymonda,*
aDepartment of Chemistry and Biochemistry, University of Berne, Freiestrasse 3, 3012 Berne, Switzerland
bChemistry Department, The Royal Veterinary and Agricultural University, Thorvaldsensvej 40, 1871 Frederiksberg C, Denmark
Received 6 February 2003; revised 2 May 2003; accepted 16 May 2003
Abstract—A series of oligo-peptide based catalysts were prepared using Fmoc solid-phase peptide synthesis. It was found that
peptides with N-terminal proline residues catalyzed an aldol reaction yielding enantiomeric enriched product. Peptide H-Pro-Glu-
Leu-Phe-OH catalyzed the reaction with good activity and moderate enantioselectivity (66% ee). Furthermore, it was shown that
an acidic side chain and/or C-termini are essential to catalysis.
# 2003 Elsevier Ltd. All rights reserved.
The aldol reaction is one of the most powerful carbon–
carbon bond forming reactions. The addition of a
nucleophilic ketone donor to an electrophilic aldehyde
acceptor in the presence of a chiral catalyst gives access
to enantiomeric rich aldol products.1
metal for aldehyde activation and a Brønsted base for
enolate generation to form an active complex.11
In this letter we report the discovery of unprotected
peptide-based catalysts for the direct asymmetric aldol
reaction. Peptides—being composed of the same build-
ing blocks (amino acid subunits) as enzymes—have the
potential to create a chiral binding environment in
which the donor is activated and reacts with the incoming
acceptor in such a way that enantiomerically enriched
aldol products could result. Side-chain protected peptides
have been reported to act as catalysts for kinetic reso-
lutions of alcohols,12 asymmetric conjugate additions of
azides13 and asymmetric phosphorylations.14 Catalytic
peptides bearing a catalytic lysine residue have been
isolated by phage display selection for binding against
1,3-diketones.15 By contrast, the peptides described here
are based on proline-type catalysis.
The direct asymmetric Aldol reaction is particularly
attractive because the energetically demanding activa-
tion of the ketone moiety to a reactive enolate-type
species is not necessary. Inspired by Nature’s aldolase
enzymes researchers have concentrated on two classes of
chiral aldolase catalysts: Type I and type II aldolase
mimics.1 Type I mimics consists most often of amino
acid catalysts, which activate the donor via enamine
formation and the acceptor through a hydrogen bond
with an acid functionality.2 Proline is the catalyst of
choice for a wide range of reactions such as Robinson
annulations,3 intermolecular aldol condensations,4
Mannich reactions,5 Michael additions,6 cyclodehy-
drations7 and a-amination of aldehydes.8 Type I aldo-
lase catalytic antibodies were obtained from designed
haptens using either a cofactor strategy9 or by reactive
immunisation against 1,3-diketone haptens.10 Key to the
function of these catalysts is an active site primary amine
functional group buried in a hydrophobic pocket with a
highly perturbed pKa amine. Type II aldolase mimics
consists of bimetallic catalysts containing a Lewis acidic
Design Principles and Synthesis
We prepared two classes of peptides following two dif-
ferent designs. The first class was based on a primary
amine as catalytic group, as seen in type I aldolases and
catalytic antibodies. This primary amine was placed in
b-position to a quaternary ammonium group, which
would lower it’s pKa down to pKa=7 as seen in type I
aldolases,10b and provide hydrophobic groups that
might enable substrate binding interactions (Class I–a).
Alternatively, we also looked at the possibility to use the
*Corresponding author. Tel.: +41-316-314-325; fax: +41-316-318-
057; e-mail: jean-louis.reymond@ioc.unibe.ch
0960-894X/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00498-0

2446
J. Kofoed et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2445–2447
peptide’s free N-terminus as a catalytic group (Class I–b)
(Fig. 1).
The peptides were first assayed for retroaldolization
catalysis in aqueous buffer based using individual stereo-
isomers of the fluorogenic aldol 1, which undergoes retro-
aldolization to form aldehyde 2 and subsequently
umbelliferone 3 by b-elimination.19 There was no activity
with any of the peptides under the conditions of this
assay (100 mMfluorogenic aldolase substrates (( R,R)-1,
(S,S)-1, (R,S)-1 and (S,R)-1), 20 mMphosphate buffer
(pH=8.00) and 10, 100 or 500 mMcatalysts). Proline
itself also did not show any measurable activity under
these conditions (Fig. 2).
The second class was based on the chemical proline-
based catalysts, and comprised peptides with either a
secondary amine (Class II–a) or proline (Class II–b) at
the N-terminus. These peptides incorporated at least
one free carboxyl function such as to compensate for
the proline’s own carboxyl group, which was used for
the peptide linkage. Proline’s free carboxyl group has
been shown to be essential for aldol-type catalysis.4a
The carboxylic function in our peptides was placed in
close proximity to the catalytic amine either directly as
the closest substituent or spatially by intercalating the
b-turn inducing aminoisobutyric acid residue next to the
proline (Fig. 1).16
We next turned our attention to the much studied con-
densation between acetone and p-nitrobenzaldehyde 4,
which has been used broadly to test proline catalysis.4a,6
The peptides, used as trifluoroacetate salts were stirred in
DMSO/acetone (4:1) with one equivalent of N-methyl
morpholine (NMM, which it itself did not catalyze the
reaction20) as buffer, and after 15 min 4 was added. The
reaction was run for 18 h and followed by isocratic RP-
HPLC. Any enantiomeric excess of the aldol product 5
was determined by chiral-phase HPLC (Fig. 3).
Class I–a peptides were obtained by reacting the Fmoc
deprotected peptidyl-resin with a large excess of alky-
lation agent (R1X, Scheme 1). Class I–b and II–b were
readily available using standard Fmoc SPPS. Class II–a
.
was obtained via CsOH H2O promoted mono-N-alkyl-
ation.17 Chloroacetylation followed by nucleophilic
displacement with amines18 (e.g., R2R3NH) gave also
access to class II–a peptides. The peptide products were
cleaved from the solid support using TFA/TIS/H2O
(95:2.5:2.5). Most peptides were sufficiently pure
Results and Discussion
Peptides containing either a primary amino group
(Class I) or an acyclic secondary amine (Class II–a)
did not catalyze the reaction (entries 1, 2 and 3, Table
1). Only N-terminal prolyl-peptides bearing an addi-
tional carboxyl group showed activity, showing that
catalytic functionality of proline’s carboxyl group
could indeed be restored by a more remote carboxyl
group, such as in the simple dipeptide H-Pro-Gly-OH
(entry 7), which also showed a moderate yet sig-
nificant enantiomeric excess in the reaction. Surpris-
ingly, the dipeptide H-Pro-Leu-OH (entry 5) was
1
(>95%) as judged by HPLC and H NMR analysis to
be used directly for screening. The remaining peptides
were further purified by preparative RP-HPLC pur-
ification before use.
catalytically inactive, suggesting
a non-productive
interaction of the hydrophobic side chain on catalysis.
Presentation of the carboxyl group as a side chain in the
dipeptide H-Pro-Asp-NH2 (entry 6) was ineffective both
in terms of catalysis and in terms of enantioselectivity.
Figure 1. Design of potential catalysts for the direct asymmetric aldol
reaction. R1 represents a hydrophobic group, R2 any amino acid side-
chain and R3 aliphatic or aromatic functionality.
Figure 2. Principle of fluorogenic aldolase assay.
.
Scheme 1. Synthesis of N-terminal modified peptides: (a) CsOH H2O,
BrCH2COOC(CH3)3, DMF; (b) R1X, DMF, DIEA; (c)
(ClCH2CO)2O, DMF/collidine (8:2); (d) R2NH2, DIEA, DMF; (e)
TFA/TIS/H2O (95:2.5:2.5).
Figure 3. Direct asymmetric aldol condensation between acetone and
p-nitrobenzaldehyde.

J. Kofoed et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2445–2447
2447
Table 1. Direct asymmetric aldol reaction between acetone and p-
nitrobenzaldehyde catalyzed by simple peptides and derivatives
References and Notes
1. Machajewski, T. D.; Wong, C.-H. Angew. Chem. Int. Ed.
2000, 39, 1352.
2. Agami, C.; Puchot, C.; Sevestre, H. Tetrahedron Lett. 1986,
27, 1501.
3. (a) Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39,
1615. (b) Eder, U.; Sauer, G.; Wiechert, R. Angew. Chem., Int.
Ed. Engl. 1971, 10, 496. (c) Agami, C.; Platzer, N.; Sevestre, H.
Bull. Soc. Chim. Fr. 1987, 2, 358.
4. (a) List, B.; Lerner, R. A.; Barbas, C. F., III J. Am. Chem.
Soc. 2000, 122, 2395. (b) Northrup, A. B.; MacMillan, D. W. C.
J. Am. Chem. Soc. 2002, 124, 6798.
5. Hanessian, S.; Pham, V. Org. Lett. 2000, 2, 2975.
6. List, B. J. Am. Chem. Soc. 2000, 122, 9336.
7. List, B.; Lerner, R. A.; Barbas, C. F., III J. Am. Chem. Soc.
1999, 120, 8131.
8. List, B. J. Am. Chem. Soc. 2002, 124, 5656.
9. (a) Reymond, J.-L.; Chen, Y. Tetrahedron Lett. 1995, 36,
2575. (b) Reymond, J.-L. Chen J. Org. Chem. 1995, 60, 6970.
10. (a) Wagner, J.; Lerner, R. A.; Barbas, C. F., III Science
1995, 270, 1797. (b) Barbas, C. F., III; Heine, A.; Zhong, G.;
Entry Class
Peptide
Conversiona eeb
1
2
I-a
I-b
5 Peptides, R1=Me
3 Peptides, R2=Ser, Phe or Leu
<5%
<5%
<5%
<5%
<5%
39%
99%
96%
94%
95%
97%
—
—
—
—
3
II-a 2 Peptides, R2=Pro, or R3=C6H11
4
5
6
7
8
9
10
11
II-b
II-b
II-b
II-b
II-b
II-b
II-b
II-b
H-Pro-Leu-NH2
H-Pro-Leu-OH
H-Pro-Asp-NH2
—
<5%
46%
66%
37%
50%
12%
H-Pro-Gly-OH
H-Pro-Glu-Leu-Phe-OH
H-Pro-Aib-Glu-Phe-OH
H-Pro-Asp-Leu-Phe-OH
H-Pro-Aib-Asp-Phe-OH
aDetermined by analytical RP-HPLC, 254 nm, on a Vydac peptide &
protein column.
bDetermined by analytical chiral HPLC, 254 nm, on a Chiralpak AS
column. The major enantiomer was assigned to be (R) by comparison
with the literature. See also ref 6.
Remote attachment of a carboxyl group via the b-turn
inducer amino-isobutyric acid afforded a catalytically
active peptide, but without improvement of enantio-
selectivity. The b-turn induction turned out to be non-
essential. Indeed, appendage with the hydrophobic pair
Leu-Phe gave the best results and restored both activity
and enantioselectivity to levels comparable to that of
proline itself, as seen with H-Pro-Asp-Leu-Phe-OH
(entry 10) and H-Pro-Glu-Leu-Phe-OH (entry 8).
Hoffmann, T.; Gramatikova, S.; Bjornestedt, R.; List, B.;
¨
Anderson, J.; Stura, E. A.; Wilson, I. A.; Lerner, R. A. Science
1997, 278, 2085. (c) Zhong, G.; Lerner, R. A.; Barbas, C. F.,
III Angew. Chem. Int. Ed. 1999, 38, 3738.
11. (a) Yamada, Y. M. A.; Yoshika, N.; Sasai, H.; Shibasaki,
M. Angew. Chem. Int. Ed. 1997, 36, 1871. (b) Trost, B. M.; Ito,
H. J. Am. Chem. Soc. 2000, 122, 12003. (c) Trost, B. M.; Ito,
H.; Silcoff, E. R. J. Am. Chem. Soc. 2001, 123, 3367.
(d) See also ref 1.
12. (a) Copeland, G. T.; Jarvo, E. R.; Miller, S. J. J. Org.
Chem. 1998, 63, 6784. (b) Vasbinder, M. M.; Elizabeth, R. J.;
Miller, S. J. Angew. Chem. Int. Ed. 2001, 40, 2824.
13. (a) Hortstmann, T. E.; Guerin, D. J.; Miller, S. J. Angew.
Chem. Int. Ed. 2000, 39, 3635. (b) Hortstmann, T. E.; Guerin,
D. J.; Miller, S. J. Org. Lett. 1999, 1, 1247.
14. Sculimbrene, B. R.; Miller, S. J. J. Am. Chem. Soc. 2001,
123, 10125.
15. Tanaka, F.; Barbas, C. F., III Chem. Commun. 2001, 769.
16. Blank, J. T.; Guerin, D. J.; Miller, S. J. J. Am. Chem. Soc.
2000, 2, 1247.
In summary, this is the first report of peptides with N-
terminal proline residues as asymmetric aldol reaction
catalysts. The catalytic activity and enantioselectivity of
proline could be matched by appendage of simple amino
acids presenting a free carboxyl function as co-catalyst.
Considered that most modifications on proline, in parti-
cular amidation of it’s carboxyl side chain,4a are incom-
patible with catalysis, our discovery opens the way for
the preparation of a large family of proline-based aldol
catalysts by standard combinatorial peptide synthesis.
17. Salvatore, R. N.; Nagle, A. S.; Kung, K. W. J. Org. Chem.
2002, 67, 674.
18. Lohne, A.; Jensen, K. B.; Lundgren, K.; Bols, M. Bioorg.
Med. Chem. 1999, 7, 1965.
Acknowledgements
19. Carlon, R. P.; Jourdain, N.; Reymond, J.-L. Chem. Eur. J.
2002, 6, 4154.
20. Tertiary amines do not catalyze the aldol reaction. See
also ref 9.
This work was supported by The Swiss National Science
Foundation, the COST program and the Office Federal
Suisse de la Recherche Scientifique.