Enantioselective aldol reaction catalysed by polyleucines

Article abstract of DOI:10.1016/j.tetasy.2007.05.024

Polyleucines of various lengths act as enantioselective catalysts in the aldol condensation between cyclohexanone and a series of aromatic aldehydes, a reaction which may be of prebiotic significance.

Full text of DOI:10.1016/j.tetasy.2007.05.024

Tetrahedron: Asymmetry 18 (2007) 1265–1268
Enantioselective aldol reaction catalysed by polyleucines
Giacomo Carrea,a Gianluca Ottolina,a Antonio Lazcano,b Vincenza Pirontic
and Stefano Colonnac,*
aIstituto di Chimica del Riconoscimento Molecolare, CNR, via Mario Bianco 9, 20131 Milano, Italy
bFacultad de Ciencias, UNAM, Apdo Postal 70-407, Cd. Universitaria, 04510 D.F., Mexico
`
Istituto di Chimica Organica ‘Alessandro Marchesini’, Facolta di Farmacia, via Venezian 21, 20133 Milano, Italy
c
Received 16 May 2007; accepted 22 May 2007
Available online 20 June 2007
Abstract—Polyleucines of various lengths act as enantioselective catalysts in the aldol condensation between cyclohexanone and a series
of aromatic aldehydes, a reaction which may be of prebiotic significance.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
These reports are consistent with our results on the behav-
iour of polyamino acids, such as L-polyleucine (PLL),
which are known to act as synthetic enzymes in the asym-
The aldol reaction, which forms C–C bonds, is a key reac-
tion in organic synthesis, with remarkable biological and
prebiological significance.1 Following the seminal experi-
ments by List, Lerner and Barbas in polar organic solvents,
many asymmetric aldol reactions have been reported in the
literature.2 Simple a-amino acids are known to catalyse not
only intramolecular, but also intermolecular aldol reac-
tions with high enantioselectivity and high yield in organic
solvent reaction media3 and in aqueous solutions,2a which
may be of prebiotic significance.3a These experiments dem-
onstrated the role of (S)-proline in promoting asymmetric
aldol reactions, as well as the catalytic asymmetric effect
of non-racemic alanine and valine in a water-based prebi-
otic carbohydrate synthesis from glycoaldehyde and form-
aldehyde.4 These results were extended with the report of
the efficient peptide-catalysed stereoselective synthesis of
tetroses by aldol condensation of glycolaldehyde.5 More-
over, Cordova et al. have recently reported that small pep-
tides catalyse the asymmetric condensations with up to
99% ee, that is, with stereoselectivities comparable with
those of natural enzymes.6 The possibility that these results
are also of prebiotic significance has been raised.7 The
thermodynamic control of asymmetric amplification based
on the equilibrium solid–liquid phase behaviour of amino
acids in solution has been described.3
metric epoxidation of chalcone and other electron-deficient
8
`
alkenes (the Julia–Colonna reaction). The kinetics and
mechanism of this reaction, which have been investigated
using a soluble PEG–polyleucine conjugate, exhibited satu-
ration kinetics for both chalcone and hydrogen peroxide,
indicating that polyleucine has an enzyme-like behaviour.9
Enantioselective catalysis was achieved with peptides with
as few as five residues while scalemic catalysts showed high
chiral amplification.10 On the basis of these results, we
speculated that simple compounds such as leucine and
small oligopeptides derived from it, such as polyleucine,
could have played a role in the enantioselective formation
of carbon–carbon bonds, not available by other abiotic
mechanisms. Leucine, together with alanine and other
amino acids of the pyruvate family, has been obtained on
mildly reducing mixtures 11; on the other hand it is likely
that polyleucine and leucine-containing oligomers were
present in the prebiotic environment.12 Following this
approach, we decided to investigate the role of leucine
and polyleucine of various lengths as potential synthetic
enzymes in the aldol condensation reactions.
2. Results and discussion
We report here the organocatalysed cross aldol reaction be-
tween cyclohexanone and aromatic aldehydes promoted by
sub-molar quantities of alanine, leucine and several leucine
*
Corresponding author. Tel.: +39 02 5031 4473; fax: +39 02 5031 4476;
e-mail: stefano.colonna@unimi.it
0957-4166/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.05.024

1266
G. Carrea et al. / Tetrahedron: Asymmetry 18 (2007) 1265–1268
OH
O
O
OH
O
O
PLL
DMSO H2O
Cl
Cl
Cl
Cl
H
syn-(2R,1'R)-1
OH
syn-(2S,1'S)-1
+
Cl
O
OH
O
anti-(2S,1'R)-1
anti-(2R,1'S)-1
Scheme 1. Aldol condensation of 4-chlorobenzaldehyde with cyclohexanone.
oligomers of different lengths in DMSO/H2O as solvent.13
The 4-chlorobenzaldehyde was chosen as a model substrate
(Scheme 1), but the reaction with four other aromatic alde-
hydes (see below) has also been investigated. L-Leucine and
L-alanine have been employed for the sake of comparison.
When cyclohexanone is the source of the enolate, as in the
present work, an E-enolate is formed, and it gives as
expected predominance of the anti products.14
Aesar was used (Table 1, line 5), the prevailing anti
diastereoisomer had the (2S,10R) absolute configuration,
that is, the configuration opposite to that found with
PLL-Bayer, whereas both the dr and the ee were practically
the same. The decrease in chemical yield can be explained
by the higher degree of polymerisation of the catalyst,
which contains 64 leucine units on an average. The higher
degree of polymerisation might affect the topological prop-
erties of the PLL-Alfa Aesar catalyst, thus affording com-
pound anti-1 with opposite configuration with respect to
the other catalysts. The PEG–polyleucine conjugate
(PLL–PEG) was inactive, in contrast with the excellent
enantioselectivities and chemical yield in the epoxidation
reaction. L-Leucine and L-alanine competed favourably
with both PLLs in terms of chemical yield, dr and ee (Table
1, lines 1 and 2) and the prevailing anti diastereoisomer
again had the (2R,10S)-absolute configuration. DMSO/
H2O in a 20:1 ratio was the solvent of choice with L-leucine
as catalyst; no reaction was observed in THF/H2O and
dioxane/H2O in a 20/1 ratio. The presence of tetrabutyl
ammonium sulphate as a phase transfer catalyst (PTC)
(Table 1, line 7) increased the yield but decreased the dr
and ee values. As a first step towards the better understand-
ing of the catalytic role of small peptides in the reactions
involving keto-bearing compounds, we extended our study
to other related aromatic aldehydes and to a-methyl cinna-
maldeyde (Scheme 2).
As shown in Table 1, antialdols were formed preferentially
in all cases, albeit with low yields. Enantioselectivity was
high for the leucine-catalysed reaction, but lower values
were observed for the different polyleucines. The results
we observed are comparable with previous reports.3a,5,7a
`
They extend the repertoire of the Julia–Colonna reaction
as a possible model of a prebiotic reaction, but also raise
a number of issues that require further study. PLL-Bayer,15
which contains seven to eight leucine residues on average,
gave the best results in terms of chemical yield, enantio-
and diastereoselectivies. The anti/syn diastereomeric ratio,
1
determined independently by H NMR and chiral HPLC,
was 5/1, while the prevailing diastereoisomer with the
(2R,10S) absolute configuration16 was obtained in 54% ee
(Table 1, line 3). The further addition of 10 mol % equiv
of PLL-Bayer (Table 1, line 4) led to a reasonable increase
both in enantioselectivity (65% ee) and in diastereoselectiv-
ity (7/1 dr). Unexpectedly, when commercial PLL-Alfa
O
OH
O
OH
Table 1. Aldol condensation of 4-chlorobenzaldehyde with cyclohexanone
catalysed by amino acids and L-polyleucine (PLL)
R
R
O
O
Catalyst
Reaction Yield
dre
eee anti eee syn
PLL
DMSO H2O
syn-(2R,1'R)-2-5
OH
syn-(2S,1'S)-2-5
OH
time (h)
(%) (anti/syn) (%)
(%)
+
H
R
L-Alaninea
96
144
120
240
168
11 days
44
35
28
33
9
18:1
30:1
5:1
7:1
3:1
92
95
54
65
48
33
21
19
6
O
O
L-Leucinea
R
R
PLL-Bayerb
PLL-Bayerc
PLL-Alfa Aesarb
PLL–PEGb
15
anti-(2S,1'R)-2-5
anti-(2R,1'S)-2-5
0
79
L-Leucine + PTCd 168
4:1
83
15
Scheme 2. For R see Table 2.
a Reaction conditions: cyclohexanone (1.5 mmol), 4-chlorobenzaldehyde
(0.5 mmol), amino acid catalyst (0.1 mmol) in DMSO/H2O (2.0 mL/
0.1 mL).
Moderate to high enantio- and diastereoselecivities were
obtained with L-leucine as a catalyst with benzaldehyde
(Table 2, entry 1), 1-naphthaldehyde (Table 2, entry 4)
and a-methyl cinnamaldehyde (Table 2, entry 8). The enan-
tio- and diastereoselectivities with PLL-Baeyer and benz-
aldehyde (Table 2, entry 2), 1-naphthaldehyde (Table
2, entry 5), 4-nitrobenzaldehyde (Table 2, entry 6) and
b PLL catalyst (0.05 mmol) was used.
c PLL catalyst (0.1 mmol) was added in two aliquots.
d Amino acid catalyst (0.1 mmol) and tetrabutyl ammonium sulfate
(0.06 mmol) were used.
e Determined by chiral phase HPLC analysis using an AD Chiralcel col-
umn, which base-line separated the four stereoisomers.

G. Carrea et al. / Tetrahedron: Asymmetry 18 (2007) 1265–1268
1267
Table 2. Aldol condensation of various aldehydes with cyclohexanone in the presence of PLLs
Entry
R
Product Catalyst
Reaction time (h) Yield (%) drc (anti/syn) eec (%) anti
eec (%) syn
1
2
3
4
5
6
7
8
9
–Ph
–Ph
–Ph
2
2
2
3
3
4
4
5
5
L-Leucinea
144
120
168
144
168
168
168
144
168
85
7
4
28
7
34
17
15
13
14:1
8:1
3:1
16:1
23:1
6:1
5:1
7:1
19:1
94 (2S,10R) 22 (2S,10S)
40 (2S,10R) 33 (2R,10R)
56 (2R,10S) 37 (2R,10R)
PLL-Bayerb
PLL-Alfa Aesarb
L-Leucinea
-1-Naphthyl
-1-Naphthyl
–Ph-4-NO2
–Ph-4-NO2
-3-Phenyl-2-propen-2-yl
-3-Phenyl-2-propen-2-yl
P99 (2R,10S)
5
PLL-Bayerb
PLL-Bayerb
PLL-Alfa Aesarb
L-Leucinea
59 (2S,10R) 19
21 (2S,10R) 34
71 (2S,10R) 47
74 (2S,10R) 29
69 (2S,10R) 45
PLL-Bayerb
a Reaction conditions: cyclohexanone (1.5 mmol), aldehyde (0.5 mmol), amino acid catalyst (0.1 mmol) in DMSO/H2O (2.0 mL/0.1 mL).
b PLL catalyst (0.05 mmol) was used.
c Determined by chiral phase HPLC analysis using an AD/OJ Chiralcel column.
a-methyl cinnamaldehyde (Table 2, entry 9) were of the same
order of magnitude as those observed with model substrate
1. In all cases, the prevailing anti-diastereisomer had the
(2S,10R)-absolute configuration. With a-methyl cinnamal-
dehyde, the results with PLL-Bayer were comparable (yield
and enantioselectivity) or even higher (diastereoselectivity)
than those obtained with leucine (Table 2, entries 8 and 9).
PLL-Alfa Aesar gave better results in terms of enantiose-
lectivity with respects to PLL-Bayer with benzaldehyde
(Table 2, entries 2 and 3) and 4-nitrobenzaldehyde (entries
6 and 7, Table 2). Surprisingly, with PLL-Alfa-Aesar the
replacement of a hydrogen atom with a nitro group gave
the opposite stereochemical course, that is, (2R,10S) for 2
(Table 2, entry 3) and (2S,10R) with aldol 4 (Table 2, entry
7). In order to ascertain the possible role of the lengthy
polypeptides in chiral amplification, scalemic polyleucine
was also explored. As shown in Table 3, in contrast with
the amplification effect observed in the asymmetric epoxi-
dation of chalcone with hydrogen peroxide in alkaline
medium,10 the effect of PLL on the aldol condensation of
cyclohexanone with 4-chlorobenzaldehyde was negligible.
This may be due to the higher degree of polymerisation
of the catalyst, which is formed by 64 leucine units on aver-
age. Work is currently in progress which will address this
issue. Interestingly non-linear effects in the acyclic amino
acid catalysed aldol reaction of cyclohexanone with
4-nitrobenzaldehyde in heterogeneous wet DMSO have
recently been reported.17 The prebiotic availability of the
reactants used in our model experiments may be highly
speculative. By the same token, it can be argued that the
DMSO/water medium employed here and by others3a is
not compatible with prebiotic conditions. However, the re-
sults reported herein suggest the existence of facile routes
to homochirality that may help to understand the enantio-
selective formation of carbon–carbon bonds.
3. Conclusions
The results presented herein show the catalytic role of var-
ious polyleucines (and of simple amino acids such as L-ala-
nine and L-leucine) in the enantioselective intermolecular
aldol condensation between cyclohexanone and a series
of aromatic aldehydes. They extend previous observations
on the catalytic role of amino acids, and strengthen the
hypothesis that these and other simple compounds (includ-
ing oligopeptides) could play a key role in prebiotic
catalysis.
References
1. Modern Aldol Reactions; Mahrwald, R., Ed.; Wiley-VCH:
Weinheim, 2004; Vols. 1,2.
2. (a) Hayashi, Y. Angew. Chem., Int. Ed. 2006, 45, 8103–8105,
and references cited therein; (b) Hayashi, Y.; Aratake, S.;
Itoh, T.; Okano, T.; Sumiya, T.; Shoji, M. Chem. Commun.
2007, 957–959.
3. (a) Klussmann, M.; White, A. J. R.; Armstrong, A.;
Blackmond, D. G. Angew. Chem., Int. Ed. 2006, 45, 7985–
7989; (b) Klussmann, M.; Iwamura, H.; Matthew, S.; Wells,
D.; Pandya, U.; Armstrong, A.; Blackmond, D. G. Nature
2006, 441, 621–623.
4. Pizzarello, S.; Weber, A. L. Science 2004, 303, 1151.
5. Weber, A. L.; Pizzarello, S. Proc. Natl. Acad. Sci. U.S.A.
2006, 103, 12713–12717.
´
6. Cordova, A.; Zou, W.; Dziedzic, P.; Ibrahem, I.; Reyes, E.;
Xu, Y. Chem. Eur. J. 2006, 12, 5383–5397.
7. (a) Cordova, A.; Ibrahem, I.; Casas, J.; Sunden, H.; Engquist,
M.; Reyes, E. Chem. Eur. J. 2005, 11, 4772–4786; (b)
Cordova, A.; Engquist, M.; Ibrahem, I.; Casas, J. J. Am.
Chem. Soc. 2004, 126, 8914–8918.
Table 3. Aldol condensation of 4-chlorobenzaldehyde with cyclohexanone
in the presence of PLLs
Catalysta (%)
Reaction Yield drb
eeb anti eeb syn
(%) (%)
´
8. Julia, S.; Masana, J.; Vega, J. C. Angew. Chem., Int. Ed. Engl.
´
1980, 19, 929–931; Julia, S.; Guixer, J.; Masana, J.; Rocas, J.;
time (h)
(%)
(anti/syn)
Colonna, S.; Annuziata, R.; Molinari, H. J. Chem. Soc.,
Perkin Trans. 1 1982, 1317–1321; Porter, M. J.; Skidmore, J.
Chem. Commun. 2000, 1215–1217; Gerlach, A.; Geller, T.
Adv. Synth. Catal. 2004, 346, 1247–1249.
D-PLL ee 9.1
D-PLL ee 20
L-PLL ee 20
168
168
168
0
2
7
1.8:1
1.8:1
1.3:1
11
7
4.5
12
7
14
L-PLL ee 42.9 168
10
9. Carrea, G.; Colonna, S.; Meek, A. D.; Ottolina, G.; Roberts,
S. M. Chem. Commun. 2004, 1412–1413; Carrea, G.; Col-
onna, S.; Kelly, D. R.; Lazcano, A.; Ottolina, G.; Roberts, S.
M. Trends Biotechnol. 2005, 23, 507–513.
a Reaction conditions: cyclohexanone (1.5 mmol), 4-chlorobenzaldehyde
(0.5 mmol), PLL catalyst (0.05 mmol) in DMSO/H2O (2.0 mL/0.1 mL).
b Determined by chiral phase HPLC analysis using AD Chiralcel column.

1268
G. Carrea et al. / Tetrahedron: Asymmetry 18 (2007) 1265–1268
10. Kelly, D. R.; Meek, A.; Roberts, S. M. Chem.Commun. 2004,
2021–2022.
11. Miller, S.; Lazcano, A. In Life’s Origin: The Beginnings of
Biological Evolution; Schopf, J. W., Ed.; California University
Press: Berkeley, 2002; p 78.
12. Leman, L.; Orgel, L.; Ghadiri, M. R. Science 2004, 306, 283–
286; Biron, J. P.; Parkes, A. L.; Pascal, R.; Sutherland, J. D.
Angew. Chem., Int. Ed. 2005, 44, 6731–6734, and references
cited therein.
13. Typical procedure for aldol condensation catalysed by PLLs.
The appropriate aldehyde (0.5 mmol) was dissolved in a
mixture of DMSO/H2O (2 mL/0.1 mL), then cyclohexanone
(1.5 mmol) after which the PLLs (see Tables 1–3) were added.
The reaction mixture was stirred at 250 rpm at room
temperature for the indicated time (see Tables 1–3). The
reaction progress was monitored by TLC using petroleum
ether/ethyl acetate (8/2) as mobile phase. After the appro-
priate reaction time, the mixture was filtered through a silica
cake to remove DMSO/H2O mixture, the filter was washed
with petroleum ether/ethyl acetate (9/1, 100 mL) and (8/2,
100 mL). The recovered organic layer was concentrated under
reduced pressure. The crude was purified by flash chroma-
tography using petroleum ether/ethyl acetate (8/2) as eluant.
The isolated products were characterised according to 1H
NMR, mass analysis and the collected data were in accor-
dance to those known in the literature. For compounds 1, 2, 4
see Ref. 16; for compounds 3 and 5; see Denmark, S. E.;
Stavenger, R. A.; Wong, K. T.; Su, X. J. Am. Chem. Soc.
1999, 121, 4982–4991.
14. Zimmerman, H. E.; Traxler, M. D. J. Am. Chem. Soc. 1957,
79, 1920–1923; Heathcock, C. H.; Buse, C. T.; Kleschick, W.
A.; Pirrung, M. C.; Sohn, J. E.; Lampe, J. J. Org. Chem. 1980,
45, 1066–1081.
15. The PLL-Baeyer was prepared according to the procedure
described in Gerlach, A.; Geller, T. Adv. Synth. Catal. 2004,
346, 1247–1249.
16. The absolute configuration was attributed according to
Mase, N.; Nakai, Y.; Ohara, N.; Yoda, H.; Takabe, K.;
Tanaka, F.; Barbas, C. F., III. J. Am. Chem. Soc. 2006, 128,
734–735.
´
17. Dziedzic, P.; Zou, W.; Ibrahem, I.; Sunden, H.; Cordova, A.
Tetrahedron Lett. 2006, 47, 6657–6661.