Synthesis of proline-derived dipeptides and their catalytic enantioselective direct aldol reactions: Catalyst, solvent, additive and temperature effects

Article abstract of DOI:10.1007/s00726-009-0290-3

A series of dipeptides of l-proline-l-amino acid and l-proline-d-amino acid were synthesized to evaluate the catalytic effect for asymmetric direct aldol reactions. In the direct aldol reaction, a catalyst of l-proline-l-amino acid achieves better enantioselectivity than the corresponding l-proline-d-amino acid catalyst. Solubility of the dipeptide catalysts in the solvents is a key point for achieving a better yield of the direct aldol reaction, while hydrogen bonding of solvent does not play an important role in attaining better enantioselectivity and yield. Yield and enantioselectivity of the direct aldol reaction in water were improved by NMM and SDS additives, but the results that were done in plain DMSO were even better.

Full text of DOI:10.1007/s00726-009-0290-3

Amino Acids (2010) 38:839–845
DOI 10.1007/s00726-009-0290-3
ORIGINAL ARTICLE
Synthesis of proline-derived dipeptides and their catalytic
enantioselective direct aldol reactions: catalyst, solvent,
additive and temperature effects
Y.-H. Chen Æ P.-H. Sung Æ Kuangsen Sung
Received: 7 February 2009 / Accepted: 31 March 2009 / Published online: 17 April 2009
Ó Springer-Verlag 2009
Abstract A series of dipeptides of L-proline-L-amino acid
and L-proline-D-amino acid were synthesized to evaluate
the catalytic effect for asymmetric direct aldol reactions. In
the direct aldol reaction, a catalyst of L-proline-L-amino
acid achieves better enantioselectivity than the corre-
sponding L-proline-D-amino acid catalyst. Solubility of the
dipeptide catalysts in the solvents is a key point for
achieving a better yield of the direct aldol reaction, while
hydrogen bonding of solvent does not play an important
role in attaining better enantioselectivity and yield. Yield
and enantioselectivity of the direct aldol reaction in water
were improved by NMM and SDS additives, but the results
that were done in plain DMSO were even better.
Majewski et al. 2006) because metal-free chiral organic
catalysts are in high demand in the pharmaceutical industry.
Since aldolases are globular proteins, organic chemists
increased the peptide length of the catalyst from a single
amino acid to a dipeptide, tripeptide and oligopeptide to see
if a catalyst with a longer peptide chain works better (Tang
et al. 2004; Lei et al. 2007; Luppi et al. 2005; Zhao and Dodda
2007; Zheng et al. 2006; Shi et al. 2004; Tsogoeva and Jagtap
2004). Recently, some of the dipeptide catalysts have been
used to carry out enantioselective direct aldol reaction. Li
et al. found that dipeptide 4 catalyzed enantioselective direct
aldol reaction of 4-nitrobenzaldehyde with acetone better
than dipeptides 1, 2, 3 and 6 in DMSO with NMM base and
PGME 5000 surfactant (Shi et al. 2004). Gong et al. reported
that dipeptide ester 5 was better than dipeptide esters 8, 9 and
13 for catalyzing enantioselective direct aldol reaction of
4-nitrobenzaldehyde with hydroxyacetone in THF/H2O
(Tang et al. 2004). Tomasini et al. studied enantioselective
direct aldol reaction of isatin with acetone catalyzed by
dipeptide esters 17, 16, 9, 15, 18 or 19 at -15°C and found
that 16 worked the best (Luppi et al. 2005). Fang et al. per-
formed enantioselective direct aldol reaction of cyclohexa-
none with 4-pyridinecarbaldehyde in water by dipeptides 4,
14, 12, 7, or 20 with NMM base and PEG400 surfactant, and
20 worked the best (Lei et al. 2007). Zhao and Dodda found
that dipeptide 11 worked best among proline 22, prolinamide
21, dipeptides 10 and 11, and dipeptide ester 9 in enantio-
selective direct aldol reaction of ethyl glyoxylate with ace-
tone (Zhao and Dodda 2007).
Keywords Dipeptide Á Direct aldol reaction Á
Enantioselective reaction
Introduction
Aldolases efficiently catalyze aldol reactions of nonactivated
unprotected polyfunctionalized substrates under mild
ambient conditions to form a new carbon–carbon bond in a
stereospecific fashion. Last decade, Barbas III, List et al.
started to recreate natural aldolase enzymes with synthetic
catalysts and found that L-proline and its analogs can work as
chiral catalysts in direct aldol reactions (Notz et al. 2004; List
2004). Later on, the concept of small organic molecules as
chiral catalysts has received great attention (Noziere and
Cordova 2008; Casas et al. 2005; Ibrahem et al. 2006; Piz-
zarello and Weber 2004; Northrup and MacMillan 2002;
The dipeptides, which have been used for direct aldol
reactions are all L-proline-L-amino acid. Thus, the question
arises: will a dipeptide of L-proline-D-amino acid give a
better enantioselectivity in direct aldol reactions? In this
article, we will address this question.
Y.-H. Chen Á P.-H. Sung Á K. Sung (&)
Department of Chemistry, National Cheng Kung University,
Tainan, Taiwan, ROC
e-mail: kssung@mail.ncku.edu.tw
123

840
Y.-H. Chen et al.
HO
O
O
O
Ph
O
S
N
N
N
N
N
N
H
H
H
N
CO2H
HO2C
HO2C
HO2C
H
N
H
3
1
2
4
OH
Ph
Ph
HO
O
O
O
N
CO2H
N
CO2Me
H
H
N
CO2H
N
H
N
H
N
H
H
6
5
7
OH
O
O
O
N
CO2Me
H
N
CO2Me
N
CO2H
N
H
H
H
N
H
N
H
9
8
10
O
AcO
O
O
N
H
CO2H
N
CO2H
N
CO2Me
N
H
H
H
N
N
12
H
H
13
11
O
O
O
N
CO2Bn
N
H
CO2Bn
N
CO2H
H
H
N
N
H
N
H
H
15
14
16
O
O
O
N
CO2Bn
H
N
CO2Bn
N
N
H
CO2Bn
H
H
N
N
18
H
H
17
19
123

Synthesis of proline-derived dipeptides
841
NH
O
O
O
OH
NH2
N
CO2H
H
N
N
N
H
H
H
22
20
21
1
Materials and methods
24c: H NMR (CDCl3, 300 MHz): d 7.26–7.31 (m, 5H,
PhH), 5.05–5.18 (m, 2H, CH2), 4.46–4.50 (m, 1H, CH),
4.34–4.45 (m, 1H, CH), 3.44–3.53 (m, 2H, CH2), 1.88–2.16
(m, 5H, CH and CH2CH2), 0.85–0.89 (m, 6H, 2CH3); 13C
NMR (CDCl3, 75 MHz): d 174.33, 172.48, 172.05, 155.93,
136.02, 128.26, 127.86, 127.67, 67.32, 60.54, 60.18, 57.02,
47.21, 30.78, 28.28, 24.32, 23.33, 18.78, 17.27; IR
Chemicals
N-Cbz-L-proline was prepared according to the literature
(Tang et al. 2003). THF was dried with Na/benzophenone
and used fresh.
(CHCl3): 3,321(OH, br), 1,681(C=O) cm-1
.
General method for preparation of 24a–e
24d: 1H NMR (CDCl3, 300 MHz): d 6.89–7.32 (m, 10H,
PhH), 4.94–5.17 (m, 2H, CH2), 4.80–4.94 (m, 1H, CH),
4.31–4.37 (m, 1H, CH), 3.4–3.64 (m, 2H, CH2), 2.95–3.2
(m, 2H, CH2), 1.79–2.06 (m, 4H, CH2CH2); 13C NMR
(CDCl3, 75 MHz): d 173.58, 172.02, 155.51, 135.81,
129.10, 128.33, 128.26, 127.99, 127.74, 126.79, 67.36,
60.66, 60.22, 52.63, 47.02, 46.75, 37.17, 30.87, 29.40,
24.04, 23.08; IR (CHCl3): 3,315(OH, br), 1,686(C=O)
To a solution of 23 (498 mg, 2 mmol) in triethylamine
(0.3 mL, 2 mmol) and 4 mL of dry THF, ethyl chlorofor-
mate (0.2 mL, 2 mmol) was added dropwise over a period
of 3 min at 0°C, followed by stirring for 3 h. The reaction
mixture was added dropwise to the solution of glycine,
D-valine, L-valine, D-phenylalanine or L-phenylalanine
(2 mmol) in 4 mL of NaHCO3 (168 mg, 2 mmol) aqueous
solution at 0°C, and then the mixture was stirred for 18 h at
room temperature. The final reaction mixture was acidified
with dilute HCl aqueous solution, followed by extraction
with ethyl acetate. The organic layer was washed with
brine, dried over anhydrous Na2SO4, filtered and concen-
trated by rotary evaporator to get the crude product. The
crude product was purified by column chromatography
with mobile phase of ethyl acetate and hexane in 1:1 ratio
to get the pure product.
cm-1
.
24e: 1H NMR (CDCl3, 300 MHz): d 7.12–7.35 (m, 10H,
PhH), 4.90–5.18 (m, 2H, CH2), 4.85 (m, 1H, CH),
4.30–4.36 (m, 1H, CH), 3.20–3.40 (m, 2H, CH2), 3.00 (m,
2H, CH2), 1.6–2.18 (m, 4H, CH2CH2); 13C NMR (CDCl3,
75 MHz): d 173.64, 171.98, 171.61, 155.94, 155.31,
135.86, 135.75, 129.10, 128.32, 127.95, 127.74, 126.80,
67.37, 67.33, 60.23, 52.68, 47.13, 46.80, 37.15, 30.58,
29.46, 28.10, 24.04, 23.03; IR (CHCl3): 3,317(OH, br),
1,682(C=O) cm-1
.
24a: 1H NMR (CDCl3, 300 MHz): d 7.19–7.44 (m,
5H, PhH), 5.26 (d, J = 12.3 Hz, 1H, CH), 5.18 (d,
J = 12.3 Hz, 1H, CH), 4.45–4.55 (m, 1H, CH), 3.97–
4.12 (m, 2H, CH2), 3.58–3.67 (m, 2H, CH2), 1.96–2.26
(m, 4H, CH2CH2); 13C NMR (CDCl3, 75 MHz): d
172.31, 171.94, 155.93, 155.60, 135.83, 128.31, 128.00,
127.76, 67.46, 60.67, 60.20, 47.30, 46.93, 41.05, 30.88,
28.68, 24.23, 23.31; IR (CHCl3): 3,323 (OH, br), 1,682
General method for preparation of 25, 26, 10, 27 or 4
To each of the solution of 24a–e (200 mg) in 2 mL of
methanol, 10% Pd/C (20 mg) was added, followed by
stirring for 3 h under a hydrogen balloon. The reaction
mixture was filtered and washed with methanol, and the
filtrate was concentrated by a rotary evaporator. The crude
product was passed through a short column of silica with
water to remove residual metal, then concentrated and
dried to get the pure product.
25: [a]2D4°C = -23.8 (c = 0.44 g/100 mL, H2O); 1H
NMR (D2O, 300 MHz): d 4.45 (t, J = 8.3 Hz, 1H, CH),
3.86 (q, J = 17.3 Hz, 2H, CH2), 3.40–3.50 (m, 2H, CH2),
2.45–2.53 (m, 1H,), 2.06–2.18 (m, 4H, CH2); 13C NMR
(D2O, 75 MHz): d 175.62, 169.25, 59.63, 46.16, 42.97,
29.22, 23.55.
(C=O) cm-1
.
1
24b: H NMR (CDCl3, 300 MHz): d 7.26–7.30 (m, 5H,
PhH), 5.12 (s, 2H, CH2), 4.47–4.53 (m, 1H, CH), 4.38–4.47
(m, 1H, CH), 3.46–3.56 (m, 2H, CH2), 1.85–2.12 (m, 5H,
CH and CH2CH2), 0.81–0.90 (m, 6H, 2CH3); 13C NMR
(CDCl3, 75 MHz): d 174.01, 172.31, 155.65, 135.99,
128.29, 127.96, 127.30, 67.33, 60.78, 60.39, 56.66, 47.33,
46.91, 30.94, 30.72, 29.39, 24.08, 23.22, 18.75, 17.15; IR
(CHCl3): 3,331(OH, br), 1,681(C=O) cm-1
.
123

842
Y.-H. Chen et al.
26: [a]2D4°C = -27.9 (c = 0.34 g/100 mL, H2O); 1H
at room temperature. The HPLC conditions and analytic
data are: chiral column, chiralcel OD-RH; mobile phase,
water/acetonitrile = 7:3; flow rate, 0.4 mL/min; UV
wavelength of detector, 271 nm; retention time for R-iso-
mer (major) and S-isomer (minor), 37, 40.8 min.
NMR (D2O, 300 MHz): d 4.47 (t, J = 6.7 Hz, 1H, CH),
4.17 (d, J = 4.3 Hz, 1H, CH), 3.40–3.49 (m, 2H, CH2),
2.4–2.6 (m, 1H, CH), 2.09–2.24 (m, 4H, CH2), 0.94
(t, J = 7.3 Hz, 6H, CH3); 13C NMR (D2O, 75 MHz): d
177.55, 169.21, 60.52, 59.62, 46.28, 29.96, 29.95, 23.55,
18.55, 16.76.
10: [a]2D4°C = -30.9 (c = 0.34 g/100 mL, H2O); 1H
NMR (D2O, 300 MHz): d 4.44 (t, J = 8.4 Hz, 1H, CH),
4.05 (d, J = 5.9 Hz, 1H, CH), 3.39–3.49 (m, 2H, CH2),
2.4–2.6 (m, 1H, CH), 2.05–2.17 (m, 4H, CH2), 0.94
(t, J = 7.1 Hz, 6H, CH3); 13C NMR (D2O, 75 MHz): d
177.88, 168.77, 61.40, 59.45,46.24, 29.79, 29.42,
23.47,18.51, 17.15.
(2S)-2-[(R)-Hydroxy(4-nitrophenyl)methyl]cyclohexanone
(anti-isomer)
1H NMR (CDCl3, 300 MHz): d 8.21 (d, J = 8.6 Hz, 2H,
PhH), 7.50 (d, J = 8.6 Hz, 2H, PhH), 4.89 (dd, J = 3.0,
8.4 Hz, 1H, CH), 4.06 (d, J = 3.0 Hz, 1H, OH), 1.3–2.6
(m, 9H, CH & CH2). The ee value was determined by GC
analysis of the silylated product, which was carried out by
reaction with triethylamine and chlorotrimethylsilane at
room temperature. The GC conditions and analytic data
are: chiral column, CP-Chirasil-Dex CB; temperature,
170°C; retention time for (R,S)-isomer (minor), (S,R)-iso-
mer (major): 35.1, 36.2 min.
27: [a]2D4°C = -40 (c = 0.25 g/100 mL aq. HCl); 1H
NMR (D2O, 300 MHz): d 7.29–7.41 (m, 5H, PhH), 4.63
(dd, J = 4.2, 10.1 Hz, 1H, CH), 4.30 (t, J = 7.6 Hz, 1H,
CH), 3.28–3.38 (m, 3H, CH and CH2), 2.90 (t, J = 10 Hz,
1H, CH), 2.18–2.25 (m, 1H, CH), 1.89–1.96 (m, 1H, CH),
1.72–1.79 (m, 1H, CH),1.50–1.57 (m, 1H, CH); 13C NMR
(D2O, 75 MHz): d 177.02, 168.53, 137.34, 128.93, 128.37,
126.62, 59.44, 55.70, 46.08, 37.48, 29.64, 23.28.
(2R)-2-[(R)-Hydroxy(4-nitrophenyl)methyl]cyclohexanone
(syn-isomer)
4: [a]2D4°C = -15.4 (c = 0.13 g/100 mL aq. HCl); H
1
NMR (D2O, 300 MHz): d 7.31–7.40 (m, 10H, PhH), 4.60
(dd, J = 4.5, 10.3 Hz, 1H, CH), 4.49 (dd, J = 5.1, 9.2 Hz,
1H, CH), 4.25–4.34 (m, 2H, 2CH), 3.22–3.44 (m, 6H,
CH2), 2.90–3.24 (m, 2H, CH2), 1.5–2.5 (m, 8H, CH2); 13C
NMR (D2O, 75 MHz): d 177.44, 177.17, 168.46, 168.33,
137.47, 128.91, 128.37, 128.34, 126.60, 126.56, 59.44,
59.39, 56.70, 56.01, 46.17, 46.07, 37.58, 37.04, 29.64,
29.24, 23.44, 23.27.
1H NMR (CDCl3, 300 MHz): d 8.21 (d, J = 8.7 Hz, 2H,
PhH), 7.49 (d, J = 8.7 Hz, 2H, PhH), 5.48 (s, 1H, CH),
3.16 (d, J = 3.2 Hz, 1H, OH), 1.3–2.6 (m, 9H, CH &
CH2). The ee value was determined by GC analysis of the
silylated product, which was carried out by reaction with
triethylamine and chlorotrimethylsilane at room tempera-
ture. The GC conditions and analytic data are: chiral col-
umn, CP-Chirasil-Dex CB; temperature, 170°C; retention
time for (S,S)-isomer (minor), (R,R)-isomer (major): 39.2,
41.4 min.
General method for direct aldol reactions
To the solution of catalyst 25, 26, 10, 27 or 4 (0.09 mmol)
in 0.8 mL of solvent and 0.2 mL of acetone or cyclohex-
anone, p-nitrobenzaldehyde (45 mg, 0.3 mmol) was added,
followed by stirring at room temperature or 0°C for some
time. Then the reaction mixture was quenched with 20 mL
of ammonium chloride aqueous solution, and the mixture
was extracted with ethyl acetate. The organic layer was
washed with brine, dried over anhydrous Na2SO4, filtered
and concentrated by a rotary evaporator to get the aldol
product.
Results and discussion
Because L-proline gives good enantioselectivity in direct
aldol reactions (Notz et al. 2004; List 2004), we synthe-
sized dipeptides from L-proline, and the second amino
acids in the dipeptides were glycine, L-valine, D-valine,
L-phenylanaline or D-phenylanaline, respectively. As
shown in Scheme 1 for the preparation of the dipeptides,
we protected the secondary amine group of L-proline (22)
with a carbobenzyloxy (cbz) group by reaction with benzyl
chloroformate in aqueous NaHCO3. Then the carboxylic
acid group of the protected L-proline was activated by
reaction with ethyl chloroformate in aqueous NaHCO3,
followed by treatment with the second amino acid. After-
wards, hydrogenation of the protected dipeptides on Pd/C
eliminated the cbz protecting group from the dipeptides.
Direct aldol reaction of acetone with p-nitrobenzalde-
hyde in the presence of 30 mol% L-Pro-Gly 25 in DMSO
4-Hydroxy-4-(4-nitrophenyl)butan-2-one
1H NMR (CDCl3, 300 MHz): d 8.21 (d, J = 8.7 Hz, 2H,
PhH), 7.54 (d, J = 8.7 Hz, 2H, PhH), 5.23–5.28 (m, 1H,
CH), 3.57 (d, J = 3.4 Hz, 1H, OH), 2.82–2.90 (m, 2H,
CH2), 2.22 (s, 3H, CH3). The ee value was determined by
HPLC analysis of the acetylated product, which was car-
ried out by reaction with triethylamine and acetyl chloride
123

Synthesis of proline-derived dipeptides
843
Scheme 1
O
O
O
1.
O
Cl
N
O
Cl
OH
O
NEt3
O
O
N
H
OH
R2
R1
H2N
NaHCO3(aq)
0°C - rt
2.
OH
22
O
23
NaHCO3(aq)
O
R1
R2
OH
O
24a, 25: R1= H, R2= H
24b, 26: R1= i-Pr, R2= H
24c, 10: R1= H, R2= i-Pr
24d, 27: R1= Bn, R2= H
24e, 4: R1= H, R2= Bn
N
R1
R2
N
N
Pd/C
H2(g)
H
OH
H
N
O
O
O
H
O
25, 26, 10, 27, 4
24a~e
was optimized by reaction time and temperature. As shown
in Table 1, a better enantioselectivity was achieved at 0°C,
and a better yield was attained at a longer reaction time
(19 h) and 0°C at the expense of enantioselectivity. Thus,
the optimized condition of 0°C and 19 h was applied to the
direct aldol reaction catalyzed by 26, 10, 27 or 4. As shown
in Table 2, L-Pro-L-Phe 4 accomplished the best enanti-
oselectivity and yield in the direct aldol reaction. In addi-
tion, it was found that L-Pro-L-Phe 4 achieved better
enantioselectivity than L-Pro-D-Phe 27, and L-Pro-L-Val 10
attained better enantioselectivity than L-Pro-D-Val 26.
According to these two examples, the L-proline-L-amino
acids achieve better enantioselectivity than the L-proline-D-
amino acid. A strange result appears in Table 2: a poor
yield of the direct aldol reaction was caused by L-Pro-L-Val
10. A possible explanation for this repeatable result is poor
solubility of 10 in DMSO.
Table 1 Yields and ee values for the direct aldol reaction of acetone
with p-nitro-benzaldehyde in the presence of 30 mol% of L-Pro-Gly
25 in DMSO
Catalyst
Reaction
temperature (°C)
Reaction
time (h)
Yield
(ee value)
L-Pro-Gly 25
L-Pro-Gly 25
L-Pro-Gly 25
L-Pro-Gly 25
25
25
0
2
5
53% (59%)
80% (65%)
25% (80%)
88% (73%)
5
0
19
Table 2 Yields and ee values for the direct aldol reaction of acetone
with p-nitrobenzaldehyde in the presence of 30 mol% of dipeptide in
DMSO at 0°C for 19 h
Catalyst
Yield
(%)
ee value
(%)
L-Pro-D-Val 26
L-Pro-L-Val 10
L-Pro-D-Phe 27
L-Pro-L-Phe 4
89
32
95
98
60
71
58
82
OH
O
O
O
30 mol% of 25, 26, 10, 27, or 4
H
+
O2N
O2N
Highly hydrophilic water with hydrogen bonding capabil-
ity, however, made both yield and enantioselectivity of the
direct aldol reaction even worse, and so did water with
saturated NaCl. Besides, mixed alcohols or aqueous alco-
hols attained intermediate yield and enantioselectivity of
the direct aldol reaction. Conversely, dipolar aprotic
DMSO, which is not a hydrogen bonding donor, achieved
much better yield and enantioselectivity of the direct aldol
reaction. According to the preceding results of solvent
Because L-Pro-L-Phe 4 worked the best as a catalyst in
this direct aldol reaction, the direct aldol reaction catalyzed
by 30 mol% of 4 at 0°C for 19 h was used for solvent
optimization. Solvents with various degrees of hydrogen
bonding were tested. As shown in Table 3, methanol
worked the best among alcohols, and alcohols of higher
hydrophobicity reduced yields of the direct aldol reaction.
123

844
Y.-H. Chen et al.
did dramatically increase the yield of the direct aldol
reaction even though its enantioselectivity was limitedly
improved.
Table 3 Solvent optimization for the direct aldol reaction of acetone
with p-nitrobenzaldehyde in the presence of 30 mol% of L-Pro-L-Phe
4 in DMSO at 0°C for 19 h
Solvent
Yield
(%)
ee
value
As shown in Table 5, enantioselectivity dramatically
increased when cyclohexanone replaced acetone in the
direct aldol reaction in the presence of 30 mol% of L-Pro-L-
Phe 4 in DMSO at 25°C for 19 h. Replacement of some
DMSO with cyclohexanone in the direct aldol reaction
resulted in a decreased yield. When this direct aldol reac-
tion was run in water, conversion was very low. Addition
of 30 mol% of NMM and SDS into the reaction system in
water did increase both yield and enantioselectivity of the
direct aldol reaction, but its result is still worse than that
carried out in plain DMSO.
DMSO
98
86
22
2
82%
78%
83%
n.d.
MeOH
EtOH
i-PrOH
Water
12
11
23
30
38%
46%
48%
81%
Sat. NaCl(aq)
1:1 MeOH/water
1:1 MeOH/i-PrOH
Table 4 Influence of a tertiary amine or surfactant on the direct aldol
reaction of acetone with p-nitrobenzaldehyde in the presence of
30 mol% of L-Pro-L-Phe 4 in water at 0°C for 19 h
OH
O
Tertiary amine
Surfactant
Yield (%)
ee value
O2N
O
100 mol% NEt3
–
–
88
0.8
85
4.1
95
56
46
23%
n.d.
anti (major)
O
30 mol% of 4
28 mol% PEG400
28 mol% PEG400
30 mol% SDS
30 mol% SDS
40 mol% PEG400
40 mol% PEG400
+
H
+
100 mol% NEt3
–
26%
n.d.
OH
O
O2N
100 mol% NEt3
100 mol% NEt3
35 mol% NMM
22%
46%
68%
O2N
syn (minor)
Conclusion
optimization, hydrogen bonding does not play an important
role in achieving better enantioselectivity and yield of the
direct aldol reaction, but solubility of the dipeptide cata-
lysts in the solvents is a key point of enantioselectivity and
yield of the direct aldol reaction.
The L-proline-L-amino acid achieves better enantioselec-
tivity of the direct aldol reaction than the corresponding
L-proline-D-amino acid. Hydrogen bonding of solvent does
not play an important role in achieving better enantiose-
lectivity and yield of the direct aldol reaction, while solu-
bility of the dipeptide catalysts in the solvents is a key
point in the yield of the direct aldol reaction. Yield and
enantioselectivity of the direct aldol reaction are much
better in plain DMSO than those in other solvents even
with addition of NMM and SDS into aqueous solutions.
Because lower yield of the direct aldol reaction
probably results from poor solubility of 4 in solvents, a
tertiary amine or surfactant was added to the reaction
system in order to improve solubility of 4. As shown in
Table 4, a surfactant of PEG400 or SDS did not improve
the yield of the direct aldol reaction, but a tertiary amine
Table 5 Influence of a solvent,
Solvent
Tertiary amine
Surfactant
Yield (anti/syn)
ee (%)
for anti/syn
tertiary amine or surfactant on
the direct aldol reaction of
cyclohexanone with p-
nitrobenzaldehyde in the
presence of 30 mol% of L-Pro-
L-Phe 4 at 25°C for 19 h
DMSOa
DMSOb
DMSOc
Water
–
–
97% (75/25)
96% (72/28)
82% (80/20)
Trace
89/14
91/13
89/17
n.d.
–
–
–
–
–
–
Water
30 mol% NMM
30 mol% NMM
30 mol% NMM
–
17% (92/8)
12% (91/9)
41% (89/11)
41/10
40/12
72/42
a
Cyclohexanone:DMSO = 1:4
Water
36 mol% PEG400
30 mol% SDS
b
˚
4 A molecular sieve
Water
c
Cyclohexanone:DMSO = 4:1
123

Synthesis of proline-derived dipeptides
845
Acknowledgment This work was supported by NSC of Taiwan,
ROC (NSC 96-2113-M-006-003)
Notz W, Tanaka F, Barbas CFIII (2004) Enamine-based organoca-
talysis with proline and diamines: the development of direct
catalytic asymmetric aldol, Mannich, Michael, and Diels–Alder
reactions. Acc Chem Res 37:580–591. doi:10.1021/ar0300468
Noziere B, Cordova A (2008) A kinetic and mechanistic study of the
amino acid-catalyzed aldol condensation of acetaldehyde in
aqueous and salt solutions. J Phys Chem A 112:2827–2837. doi:
10.1021/jp7096845
Pizzarello S, Weber AL (2004) Prebiotic amino acids as asymmetric
catalysts. Science 303:1151. doi:10.1126/science.1093057
Shi L-X, Sun Q, Ge Z-M, Zhu Y-Q, Cheng T-M, Li R-T (2004)
Dipeptide-catalyzed direct asymmetric aldol reaction. Synlett
12:2215–2217
References
Casas J, Engqvist M, Ibrahem I, Kaynak B, Cordova A (2005) Direct
amino acid-catalyzed asymmetric synthesis of polyketide sugars.
Angew Chem Int Ed 44:1343–1345. doi:10.1002/anie.
200461400
Ibrahem I, Zou W, Xu Y, Cordova A (2006) Amino acid-catalyzed
asymmetric carbohydrate formation: organocatalytic one-step de
novo synthesis of keto and amino sugars. Adv Synth Catal
348:211–222. doi:10.1002/adsc.200505323
Tang Z, Jiang F, Yu L-T, Cui X, Gong L-Z, Mi A-Q, Jiang Y-Z, Wu
Y-D (2003) Novel small organic molecules for a highly
Lei M, Shi L, Li G, Chen S, Fang W, Ge Z, Cheng T, Li R (2007)
Dipeptide-catalyzed direct asymmetric aldol reactions in the
presence of water. Tetrahedron 63:7892–7898. doi:10.1016/j.tet.
2007.05.077
List B (2004) Enamine catalysis is a powerful strategy for the
catalytic generation and use of carbanion equivalents. Acc Chem
Res 37:548–557. doi:10.1021/ar0300571
enantioselective direct aldol reaction.
125:5262–5263. doi:10.1021/ja034528q
J Am Chem Soc
Tang Z, Yang Z-H, Cun L-F, Gong L-Z, Mi A-Q, Jiang Y-Z (2004)
Small peptides catalyze highly enantioselective direct aldol
reactions of aldehydes with hydroxyacetone: unprecedented
regiocontrol in aqueous media. Org Lett 6:2285–2287. doi:
10.1021/ol049141m
Luppi G, Cozzi PG, Monari M, Kaptein B, Broxterman QB, Tomasini
C (2005) Dipeptide-catalyzed asymmetric aldol condensation of
acetone with (N-alkylated) isatins. J Org Chem 70:7418–7421.
doi:10.1021/jo050257l
Majewski M, Niewczas I, Palyam N (2006) Acids as proline co-
catalysts in the aldol reaction of 1, 3-dioxan-5-ones. Synlett
15:2387–2390. doi:10.1055/s-2006-950421
Northrup AB, MacMillan DWC (2002) The first direct and enantio-
selective cross-aldol reaction of aldehydes. J Am Chem Soc
124:6798–6799. doi:10.1021/ja0262378
Tsogoeva S, Jagtap SB (2004) Dual catalyst control in the chiral
diamine-dipeptide-catalyzed asymmetric Michael addition. Syn-
lett 14:2624–2626. doi:10.1055/s-2004-834830
Zhao C-G, Dodda R (2007) Organocatalytic enantioselective synthe-
sis of secondary a-hydroxycarboxylates. Synlett 1605–1609.
doi:10.1055/s-2007-982552
Zheng J-F, Li Y-X, Zhang S-Q, Yang S-T, Wang X-M, Wang Y-Z, Bai
J, Liu F-A (2006) New N-terminal prolyl-dipeptide derivatives as
organocatalysts for direct asymmetric aldol reaction. Tetrahedron
Lett 47:7793–7796. doi:10.1016/j.tetlet.2006.08.084
123