Highly enantioselective direct syn- and anti-aldol reactions of dihydroxyacetones catalyzed by chiral primary amine catalysts

Article abstract of DOI:10.1021/ol703023t

We present herein simple primary-tertiary diamine-Bronsted acid conjugates that catalyze both syn- and anti-aldol reactions of dihydroxyacetones (DHAs) with high diastereoselectivities and enantioselectivities. This type of organocatalysts functionally mimics all four DHA aldolases, namely L-fuculose-1-phosphate aldolase, D-tagatose-1,6-diphosphate aldolase, D-fructose-1,6-diphosphate aldolase, and L-rhamnulose-1-phosphate aldolase.

Full text of DOI:10.1021/ol703023t

ORGANIC
LETTERS
2008
Vol. 10, No. 4
653-656
Highly Enantioselective Direct syn- and
anti-Aldol Reactions of
Dihydroxyacetones Catalyzed by Chiral
Primary Amine Catalysts
Sanzhong Luo,*,† Hui Xu,† Long Zhang,‡ Jiuyuan Li,† and Jin-Pei Cheng*,†,‡
Beijing National Laboratory for Molecular Sciences (BNLMS), Center for Chemical
Biology, Institute of Chemistry and Graduate School, Chinese Academy of Sciences,
Beijing 100080, China, and Department of Chemistry and State Key Laboratory of
Elemento-organic Chemistry, Nankai UniVersity, Tianjin 300071, China
luosz@iccas.ac.cn; chengjp@most.cn
Received December 17, 2007
ABSTRACT
We present herein simple primary
(DHAs) with high diastereoselectivities and enantioselectivities. This type of organocatalysts functionally mimics all four DHA aldolases,
namely -fuculose-1-phosphate aldolase, -tagatose-1,6-diphosphate aldolase, -fructose-1,6-diphosphate aldolase, and -rhamnulose-1-phosphate
aldolase.
−tertiary diamine−Brønsted acid conjugates that catalyze both syn- and anti-aldol reactions of dihydroxyacetones
L
D
D
L
Dihydroxyacetone (DHA) and its derivatives are versatile
C3-building blocks in chemical and enzymatic synthesis of
carbohydrates.1 In nature, enzymes, e.g., DHAP-dependent
aldolases, utilize dihydroxyacetone phosphate (DHAP) as a
specific donor for the synthesis of carbohydrates via aldol
coupling. These reactions generated two new stereocenters
in the form of syn- or anti-1,2-diols. Hence, there are four
types of DHAP aldolases, respectively, responsible for the
formation of the four possible diastereoisomers of 1,2-diols.
These DHAP-dependent processes have now been meticu-
lously harnessed for chemo-enzymatic synthesis of a diverse
range of stereochemically complex sugars.1b,2 On the other
hand, chemists have been long pursuing simple chemical
mimics for nature’s aldolase enzymes. However, truly
asymmetric catalysis with DHA derivatives as donors has
only been achieved very recently with the emerging orga-
nocatalysts.3 In this regard, L-proline was shown to be the
sole effective organocatalyst in most of these transformations,
and the reactions are largely limited to cyclic protected DHAs
such as 2,2-dimethyl-1,3-dioxan-5-one (2b), affording mainly
anti-aldols. Very recently, Barbas reported that primary
amino acid could catalyze the syn-aldol reation of DHA and
protected DHA.4,5 Herein, we report the first example of
small molecular catalysts that catalyzed both the syn- and
anti-aldol reactions of DHAs with high yields and excellent
stereoselectivity, thus providing one type of primary amines
† Chinese Academy of Sciences.
‡ Nankai University.
(1) (a) Enders, D.; Voith, M.; Lenzen, A. Angew. Chem., Int. Ed. 2005,
44, 1304-1325. (b) Gijsen, H. J. M.; Qiao, L.; Fitz, W.; Wong, C.-H. Chem.
ReV. 1996, 96, 443-473.
(2) Wong, C.-H.; Machajewski, T. D. Angew. Chem., Int. Ed. 2000, 39,
1352-1375.
10.1021/ol703023t CCC: $40.75
© 2008 American Chemical Society
Published on Web 01/24/2008

catalyst that functionally imitates all the four types of DHAP-
aldolases (Scheme 1).
A series of chiral primary amine catalysts were first tested
in the reactions of 2a. It was found that the primary-tertiary
diamines such as chiral trans-N,N-dialkylated diaminocy-
clohexanes 1 demonstrated activity and stereoselectivity
superior to that obtained with other primary amine catalysts
(see the Supporting Information for details). For example,
trans-N,N-dimethyldiaminocyclohexane 1a was observed to
catalyze the reaction very smoothly, affording preferentially
the syn-aldol product with 92% ee (Table 1, entry 1). Further
Scheme 1. General Strategy for the syn- and anti-Aldol
Reaction of DHAs
Table 1. Selected Screening Results
We have recently reported an organocatalytic syn-aldol
reaction of linear aliphatic ketones catalyzed by primary-
tertiary diamine-Brønsted acid catalyst.6 This reaction
occurred through a Z-enamine intermediate and thus led to
syn stereoselectivity which is distinct from that observed with
secondary amines catalysts. On the basis of the previous
results, we envisioned that a significant extension of the
reaction to acyclic DHA derivatives including free DHA may
also induce syn selectivity via Z-enamine intermediates
(Scheme 1, I). While for the reactions of cyclic DHA
derivatives, which are capable of forming only E-enamine
(Scheme 1, II), we expected that the same type of catalysts
would lead to the formation of anti-aldol products.
cat.
(mol %)
time
(h)
yieldb
(%)
eed
(%)
entrya
syn/antic
1
2
3
4
5
6
7
8
9
10
11
12
13
1a (20)
1b (20)
1c (20)
1c (20)e
1c (20)f
1d (20)
1e (20)
1e (10)g
1a (10)
1b (10)
1c (10)
1e (10)
1b (10)h
19
19
19
19
19
19
19
19
48
48
48
48
17
3a/60
3a/60
3a/91
3a/9
3a/85
3a/68
3a/99
3a/97
3b/65
3b/78
3b/47
3b/5
90:10
87:13
95:5
75:25
95:5
93:7
97:3
97:3
1:6
92
94
98
29
97
96
99
99
92
95
91
1:6
1:4
not determined
1:6
The free DHA 2a and protected DHA 2b were selected
to represent the acyclic and cyclic DHA donors, respectively.
3b/90
94
a Entries 1-7: 0.25 mmol reaction in DMSO in the presence of cat./
TfOH/m-NO2PhCOOH (1:1:1). Entries 9-12: 0.25 mmol reaction in
CH2Cl2 in the presence of Cat/TfOH/m-NO2PhCOOH (1:1:1). b Isolated
yields. c Determined by 1H NMR or chiral HPLC. d ee of major isomer,
determined by chiral HPLC. e Without TfOH and m-NO2PhCOOH. f With-
out m-NO2PhCOOH. g 10 mol % of catalyst system in DMF. h Under neat
conditions
(3) (a) Enders, D.; Grondal, C. Angew. Chem., Int. Ed. 2005, 44, 1210-
1212. (b) Suri, J. T.; Ramachary, D. B.; Barbas, C. F., III. Org. Lett. 2005,
7, 1383-1385. (c) Ibrahem, I.; Co´rdova, A. Tetrahedron Lett. 2005, 46,
3363-3367. (d) Westermann, B.; Neuhaus, C. Angew. Chem., Int. Ed. 2005,
44, 4077-4079. (e) Enders, D.; Grondal, C.; Vrettou, M.; Raabe, G. Angew.
Chem., Int. Ed. 2005, 44, 4079-4083. (f) Suri, J. T.; Mitsumori, S.;
Albertshofer, K.; Tanaka, F.; Barbas, C. F., III. J. Org. Chem. 2006, 71,
3822-3828. (g) Co´rdova, A.; Zou, W.; Dziedzic, P.; Ibrahem, I.; Reyes,
E.; Xu, Y. Chem. Eur. J. 2006, 12, 5383-5397. (h) Ibrahem, I.; Zou, W.;
Xu, Y.; Co´rdova, A. AdV. Synth. Catal. 2006, 348, 211-222. (i) Grondal,
C.; Enders, D. Tetrahedron 2006, 62, 329-337. (j) Grondal, C.; Enders,
D. AdV. Catal. Synth. 2007, 349, 694-702. For other organocatalytic de
novo syntheses of sugars, see: (k) Chowdar, N. S.; Ramachary, D. B.;
Co´rdova, A.; Barbas, C. F., III. Tetrahedron Lett. 2002, 43, 9591-9595.
(l) Northrup, A. B.; MacMillan, D. W. C. Science 2004, 305, 1752-1755.
(m) Casas, J.; Engqvist, M.; Ibrahem, I.; Kaynak, B.; Co´rdova, A. Angew.
Chem., Int. Ed. 2005, 44, 1343-1345.
(4) For direct aldol reaction of free DHA with rather low stereoselectivity
or racemic products, see ref 3g and: (a) Cordova, A.; Notz, W.; Barbas, C.
F., III. Chem. Commun. 2002, 3024-3025. (b) Market, M.; Mulzer, M.;
Schetter, B.; Mahrwald, R. J. Am. Chem. Soc. 2007, 129, 7258-7259. (c)
Kofoed, J.; Darbre, T.; Reymond, J.-L. Chem. Commun. 2006, 1482-1484.
(5) (a) Ramasastry, S. S. V.; Albertshofer, K.; Utsumi, N.; Tanaka, F.;
Barbas, C. F., III. Angew. Chem., Int. Ed. 2007, 46, 5572-5575. (b) Utsumi,
N.; Imai, M.; Tanaka, F.; Ramasastry, S. S. V.; Barbas, C. F., III. Org.
Lett. 2007, 9, 3445-3448.
screening of the analogous catalysts bearing longer alkyl
chains led to improved yields and stereoselectivity (Table
1, entries 2, 3, 6, and 7). The best results were obtained with
the N,N-didecanyl derivative 1e. Under the catalysis of 1e,
the product was isolated with a dr (syn/anti) value greater
than 30:1, an ee value greater than 99%, and a quantitative
yield (Table 1, entry 7). Consistent with our earlier observa-
tions, a strong Brønsted acid such as TfOH was essential
for the catalysis as the reaction in the absence of TfOH was
observed to give only 9% yield and poor stereoselectivity
(syn/anti ) 75:25, 29% ee, entry 4), suggesting a bifunctional
enamine catalysis rather than a tertiary amine mediated
process in our reactions.4b Addition of a second weak
Brønsted acid could further improve the yield (Table 1, entry
3 vs 5). The reaction was further optimized to use 10 mol
% of 1e in DMF and still offered good results under these
conditions (Table 1, entry 8).
(6) (a) Luo, S.; Xu, H.; Li, J.; Zhang, L.; Cheng, J.-P. J. Am. Chem.
Soc. 2007, 129, 3074-3075; For other organocatalytic syn-aldol reactions,
see: (b) Ramasastry, S. S. V.; Zhang, H.; Tanaka, F.; Barbas, C. F., III. J.
Am. Chem. Soc. 2007, 129, 288-289. (c) Kano, T.; Yamaguchi, Y.; Tanaka,
Y.; Maruoka, K. Angew. Chem., Int. Ed. 2007, 46, 1738-1740. (d) Wu,
X.; Jiang, Z.; Shen, H.-M.; Lu, Y. AdV. Synth. Catal. 2007, 349, 812-816.
654
Org. Lett., Vol. 10, No. 4, 2008

Next, the direct aldol reactions of cyclic DHA 2b were
examined. The primary-tertiary diamine catalyst 1 was again
found to be the optimal catalyst (for details, see the
Supporting Information), affording anti-aldol products. The
diamine catalysts (1a-e) with varied alkyl chains all gave
good anti diastereoselectivity and high ee values, of which
1b exhibited the best activity (Table 1, entry 10).
4). In the catalysis of 1e-TfOH-m-NO2PhCOOH (10 mol
%), reactions of the unprotected DHA 2a with a variety of
Table 4. anti-Aldol Reaction of Cyclic DHA 2b
Under optimized conditions, the reaction was completed
in 17 h in the presence of 10 mol % 1b, yielding 90% of the
anticipated products with 6:1 dr and 94% ee. In most cases,
the current catalysis demonstrated significant improvement
over previous results that required 20-30 mol % of catalyst
loading and 2-3 days of reaction time.3a-c
Table 2. syn-Aldol Reaction of Free DHA 2a
1
a 0.25 mmol reaction. b Isolated yields. c Determined by H NMR. d ee
of anti isomer, determined by chiral HPLC. e 20 mol % of catalyst was
used.
entrya
R
time (h) yieldb (%) syn/antic eed (%)
1
2
3
4
4-NO2Ph
2-NO2Ph
3-NO2Ph
4-CF3Ph
4-ClPh
1-Napth
4-CNPh
3-BrPh
Ph
19
36
36
34
60
60
40
96
84
3a/97
4a/93
5a/91
6a/60
7a/61
8a/40
9a/74
10a/50
11a/43
97:3
97:3
97:3
97:3
96:4
98:2
97:3
99:1
92:8
99
98
98
98
98
98
96
84
85
aldehyde acceptors produced the desired products in high
yields with excellent diastereo- and enantioselectivity. Sig-
nificantly, all reactions gave excellent syn diastereoselectivity
(syn/anti >30:1) and high enantioselectivity (84-99% ee).
The reaction of acyclic hydroxyacetone has also been
examined (Table 3). Though the reaction gave a low yield
in DMF, excellent syn diastereoselectivity and enantiose-
lectivity were still achieved (Table 3, entry 1).7
The switch to nonpolar solvents such as hexane led to
significantly improved yield (Table 3, entry 2), a result
ascribale to the increased reactant concention as both the
reactants and catalysts are less soluble in hexane.8 In this
case, catalyst 1c gave slightly better yield than that of 1e.
High syn selectivity was consistently obtained in the reactions
of hydroxyacetone (Table 3, entries 2-10), validating our
hypothesis (Scheme 1).
5
6e
7
8e
9e
a 0.25 mmol reaction in DMF. b Isolated yields. c Determined by chiral
HPLC. d ee of syn isomer, determined by chiral HPLC. e 20 mol % of
catalyst was used.
We next investigated the substrate scope of both syn-aldol
reactions (Tables 2 and 3) and the anti-aldol reactions (Table
In regard to the cyclic DHA donors (Table 4), the reactions
of cyclic DHA 2b in the presence of 1b generated anti-aldol
products (up to >10:1 anti/syn) with excellent enantiose-
lectivity (up to 97% ee) and high yields (up to 99%).
Aliphatic aldehydes have also been tried for both the syn-
and anti-aldol reactions but gave lower yields due to side
reactions. The catalytic system of 1b also catalyzed the
dimerization of 2b, giving a good yield but a moderate ee
(Table 4, entry 9).
Table 3. syn-Aldol Reaction of Acyclic Hydroxyacetone
entrya
R
time (h) yieldc (%) syn/antid eee (%)
1b
2
3
4
5
6
7
8
4-NO2Ph
4-NO2Ph
4-CF3Ph
4-CNPh
2-NO2Ph
4-ClPh
2-ClPh
2-BrPh
Ph
15
15
18
18
36
36
36
36
36
36
3c/25
3c/>99
4c/99
5c/88
6c/91
7c/92
8c/99
9c/92
10c/99
11c/50
93:7
94:6
88:12
91:9
99:1
88:12
89:11
94:6
94:6
94:6
96
96
91
88
99
92
95
93
91
93
We obtained the X-ray crystal structure of product 7c
(Figure 1),9 which proved the (3S,4R) configurations of the
(7) For examples on organocatalytic anti-aldol reactions of hydroxyac-
etone, see: (a) Notz, W.; List, B. J. Am. Chem. Soc. 2000, 122, 7386-
7387. (b) Sakthivel, K.; Notz, W.; Bui, T.; Barbas, C. F., III. J. Am. Chem.
Soc. 2001, 123, 5260. (c) Chen, X.-H.; Luo, S.-W.; Tang, Z.; Cun, L.-F.;
Mi, A.-Q.; Jiang, Y.-Z.; Gong, L.-Z. Chem. Eur. J. 2007, 13, 689. For syn-
aldol reaction of hydroxyacetone, see refs 5, 6, and: (d) Xu, X.-Y.; Wang,
Y.-Z.; Gong, L.-Z. Org. Lett. 2007, 9, 4247-4249.
(8) The reaction under solvent-free conditions exhibited a faster rate but
somehow lower enantioselectivity: 4 h, 91% yield, 89:11 syn/anti, 89%
ee.
9
10e
4-MeOPh
a 0.25 mmol reaction. b Reaction in DMF. c Isolated yields. d Determined
1
by H NMR. e ee of syn isomer, determined by chiral HPLC.
Org. Lett., Vol. 10, No. 4, 2008
655

diamine-Brønsted acid catalysts. Since the enantiomers of
catalysts 1 are easily accessible, this specific type of small
molecular catalysts may thus functionally mimic all the four
DHA aldolases in living systems. It is noted that the current
catalysis complements the enzymatic processes by working
selectively with aromatic aldehyde acceptors. This study also
provides one of the most efficient de novo syntheses of
sugars via organocatalysis. Further explorations along this
line are currently underway.
Figure 1. X-ray crystal structure of 7c.
Acknowledgment. This work was supported by the
Natural Science Foundation of China (NSFC 20632060,
20542007, and 20702052), the Ministry of Science and
Technology (MoST), and the Chinese Academy of Sciences.
syn product. The configurations of other syn-aldol products
can therefore be referred. The absolute configurations (as
drawn in Scheme 1) of the anti-aldol products were assigned
by HPLC and optical rotation comparisons with published
results (see the Supporting Information for details).3,5,10
To conclude, we have developed here a group of highly
efficient and stereoselective syn- and anti-aldol reactions of
dihydroxyacetones catalyzed by the simple primary-tertiary
Supporting Information Available: Experimental details
and characterizations of new products. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL703023T
(9) Crystal data for 7c: C10H11ClO3, colorless block, Mr ) 214.64, crystal
size 0.29 × 0.17 × 0.08 mm3, orthorhombic, space group P2(1)/c, a )
8.5744(17) Å, b ) 5.6867(11) Å, c ) 10.986(2) Å, R ) 90.00°, â ) 104.5°,
γ ) 90.00°, V ) 518.59(18) Å3, Z ) 2, Fcalcd ) 1.375 g cm-3, T ) 296(2)
K. A total of 1882 reflections and 1652 parameters were used for the full-
matrix, least-squares refinement on F2. R1 ) 0.0675 [I > 2σ(I)], R1 )
0.0757 (all data), wR2 ) 0.1536 [I > 2σ(I)], wR2 ) 0.1581 (all data).
(10) Enantiomers of catalysts 1, e.g., (S,S)-1b and -1e, have also been
tried in the reactions, which gave products with opposite configurations to
those of the products 3-11. Both enantiomers were compared with known
compounds by HPLC analysis. For known compounds, see ref 3f,h and:
Enders, D.; Prokopenko, O. F.; Raabe, G.; Runsink, J. Synthesis 1996, 1095.
656
Org. Lett., Vol. 10, No. 4, 2008