Mimicking fructose and rhamnulose aldolases: Organocatalytic syn-aldol reactions with unprotected dihydroxyacetone

Article abstract of DOI:10.1002/anie.200701269

(Chemical Equation Presented) Completing the quattuorvirate: Amino acids that contain primary amines directly catalyze syn-selective asymmetric aldol reactions between unprotected dihydroxy-acetone and a variety of aromatic and aliphatic aldehydes (see sc

Full text of DOI:10.1002/anie.200701269

Communications
DOI: 10.1002/anie.200701269
Asymmetric Catalysis
Mimicking Fructose and Rhamnulose Aldolases: Organocatalytic
syn-Aldol Reactions with Unprotected Dihydroxyacetone**
S. S. V. Ramasastry, Klaus Albertshofer, Naoto Utsumi, Fujie Tanaka, and Carlos F. Barbas, III*
Carbohydrates play diverse and essential roles in biology,
medicine, and industry.[1] As a result of their wide range of
uses, carbohydrates and carbohydrate-like molecules that
contain arrays of defined stereocenters substituted with
hydroxy groups have garnered significant attention from the
synthetic community.[2] One might argue that the challenge of
synthesizing defined arrays of stereocenters substituted with
hydroxy groups in an efficient diastereo- and enantiocon-
trolled fashion has been best met when chemists have
harnessed natureꢀs aldolase enzymes.[3] Our own studies
aimed at the creation of man-made aldolase enzymes,
aldolase antibodies, led us to discover the potential of
amino acids as aldolase enzyme mimics.[4] This approach has
since provided organocatalytic syntheses of carbohydrates
through aldol reactions catalyzed by proline and related
amines.[5,6] Arguably, the C3 + Cn strategy[7] is most favored by
nature for the synthesis of carbohydrates, and is facilitated by
the dihydroxyacetone phosphate family of aldolases.[3] This
family of four aldolases has been developed into unrivaled
synthetic tools that provide efficient access to carbohydrates
through the installation of 1,2-diol units in any of the four
possible diastereomeric configurations (Figure 1).[3]
Recently, our research group[6d,f] and the research group of
Enders[6c,e,g–k] have developed efficient organocatalytic carbo-
hydrate syntheses that emulate these particular aldolases and
the C3 + Cn strategy. These studies have largely used 2,2-
dimethyl-1,3-dioxan-5-one as a C3 donor in aldol reactions
catalyzed by proline or (S)-2-pyrrolidine tetrazole. Since
these catalysts provide access to anti 1,2-diols, they mimic d-
tagatose 1,6-diphosphate and l-fuculose 1-phosphate aldo-
lases. Access to syn 1,2-diols by using organocatalysis and a C3
dihydroxyacetone-based strategy has not been possible.
Recently, we reported the first syn-selective organocatalytic
aldol reactions that use unmodified a-hydroxyketones as
donors to install a syn 1,2-diol functionality in the products.[8]
Herein we disclose a significant elaboration of our strategy
with the first successful diastereo- and enantioselective
organocatalytic aldol reactions of unprotected dihydroxyace-
tone; these reactions functionally mimic those catalyzed by
the l-rhamnulose 1-phosphate and d-fructose 1,6-diphos-
phate aldolases.
Based on our earlier success in the development of syn-
aldol reactions catalyzed by amino acids that contained
primary amines,[8a] we initially screened five readily available
catalysts (Figure 2) in the aldol reaction of dihydroxyacetone
Figure 1. Substrates of the four dihydroxyacetone phosphate aldolases.
Arrows indicate the bond that is formed or broken by the action of the
2À
Figure 2. Structures of catalysts studied. TBDPS=tert-butyldiphenyl-
silyl, Tf=trifluoromethanesulfonyl.
aldolase enzyme. P=PO3
.
[*] Dr. S. S. V. Ramasastry, K. Albertshofer, Dr. N. Utsumi,
Prof. Dr. F. Tanaka, Prof. Dr. C. F. Barbas, III
The Skaggs Institute for Chemical Biology and
the Departments of Chemistry and Molecular Biology
The Scripps Research Institute
with 4-nitrobenzaldehyde using N-methylpyrrolidone (NMP)
with added water as the solvent. To facilitate determination of
the stereoselectivities of these reactions, the triol product was
peracetylated and analyzed by HPLC on a chiral stationary
phase. All the catalysts preferentially provided the desired
syn-aldol product (Table 1); however, considerations of
stereoselectivity, reaction time, and catalyst availability led
us to focus our optimization studies on the commercially
available O-tBu-l-Thr (3; Table 1, entry 3). It should be noted
that earlier studies of reactions that involved unprotected
10550 North Torrey Pines Road, La Jolla, California 92037 (USA)
Fax: (+1)858-784-2583
E-mail: carlos@scripps.edu
[**] This study was supported in part by The Skaggs Institute for
Chemical Biology.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
5572
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5572 –5575

Angewandte
Chemie
Table 1: Catalyst screening.[a]
Table 3: Additive screening for the aldol reactions catalyzed by 3.[a]
d.r.[b]
ee [%][b]
Entry
Cat.
t [days]
d.r.[b]
ee [%][b]
Entry
Additive
Solvent
t [h]
(syn/anti)
(syn/anti)
(syn/anti)
(syn/anti)
1[c]
2
3
AcOH
TFANMP
pTSANMP
PhCO2H
PhCO2H
triazole
triazole
triazole
triazole
–
NMP
17
20
7:1
6:1
87/14
1
2
3
4
5
1
2
3
4
5
5
5
0.8
3
1
1.5:1
3:1
9:1
16:1
3:1
12/24
78/34
88/30
88/78
86/50
90/48
85/56
9:1
4:1
6:1
9:1
10:1
4:1
9:1
4:1
37
9:1
30
20
30
20
48
20
20
20
20
20
16
4
NMP
NMP
NMP
NMP
DMF
DMF
NMP
DMF
NMP
NMP
DMF
85/46
91/12
92/56
90/20
87/2
95/22
88/30
92/46
94/44
95/50
92/20
5[c]
6
[a] See the Supporting Information for details. Added 3 vol% of water
with respect to NMP. py=pyridine. [b] Determined by HPLC on a chiral
stationary phase.
7[c]
8
9[c]
10[c]
11[c]
12[c]
13
14
–
9
9
9
dihydroxyacetone-based aldols and secondary amine or
primary amine catalysis failed to control the diastereoselec-
tivity of this reaction.[9] In an effort to improve reaction rates
and stereoselectivities, we evaluated different solvents and
additives. Initially, we performed a solvent screen using O-
tBu-l-Thr (3) with acetic acid as a fixed additive at 10 mol%
(Table 2). A slight enhancement was observed in reactivity
8:1
4:1
15:1
[a] See the Supporting Information for reaction conditions. TFA=tri-
fluoroacetic acid, pTSA=para-toluenesulfonic acid, triazole=1H-1,2,3-
triazole, 9=5-methyl-1H-tetrazole. [b] Determined by HPLC on a chiral
stationary phase. [c] Added 3 vol% of water with respect to NMP or
DMF.
Table 2: Effect of solvents on the aldol reaction catalyzed by O-tBu-l-Thr
(3).[a]
acceptor aldehydes provided the desired syn-aldol products in
moderate to good yield with excellent diastereo- and enan-
tioselectivity (Table 4). Significantly, O-tBu-l-Thr catalysis
provided the desired syn-aldol products with diastereo- and
enantioselectivities (up to 15:1 d.r., syn favored, and
> 99% ee, syn product) that typically exceeded those pro-
vided using l-proline catalysis.[6c–k] The use of protected
dihydroxyacetone donor 2,2-dimethyl-1,3-dioxan-5-one and
aromatic acceptor aldehydes in the anti-selective aldol
reaction also gave excellent diastereo- and enantioselectiv-
ities (though mostly less than 6:1 d.r., anti favored, and up to
94% ee). While most of the yields of the isolated product
range from modest to good, further optimization of the
methods for product isolation will likely lead to significant
improvements in yields given the challenges presented by
these free polyol products. We assigned the absolute config-
uration of products by using imidazole-catalyzed epimeriza-
tion of the corresponding 2,2-dimethyl-1,3-dioxan-5-one-
derived products obtained through l-proline catalysis (see
the Supporting Information). The 3R, 4S absolute configu-
rations of the other products shown in Table 4 were assigned
by analogy, and are in accord with our studies involving
reactions of monohydroxyketones catalyzed by 3.[8a]
Entry
Solvent[b]
t [h]
d.r.[c]
(syn/anti)
ee [%][c]
(syn/anti)
1
2
3
4
5
6
NMP
DMF
DMSO
iPrOH
CH3CN
EtOAc
16
22
170
22
165
170
6:1
8:1
8:1
4:1
4:1
6:1
92/49
94/18
80/20
66/68
90/62
94/28
[a] See the Supporting Information for reaction conditions. DMF=N,N-
dimethylformamide, DMSO=dimethyl sulfoxide. [b] Reaction in meth-
anol, trifluoroethanol, or dichloromethane was sluggish, and only a trace
of product was found after 8 days of reaction. [c] Determined by HPLC on
a chiral stationary phase.
and enantioselectivity with NMP as the solvent and acetic
acid as the additive when compared to water. A variety of
polar aprotic solvents were tolerated (Table 2). Optimal
results were obtained by using NMP/acetic acid or DMF/
acetic acid conditions (Table 2, entries 1 and 2). Reaction
optimization then focused on the acid additive component by
using either NMP or DMF as the solvent (Table 3). DMF/5-
methyl-1H-tetrazole was identified as the optimal solvent/
acid additive combination.
By using these optimized reaction conditions, we studied
the enantioselective direct organocatalytic syn-aldol reaction
of unprotected dihydroxyacetone. Reaction of unprotected
dihydroxyacetone with a variety of aromatic and aliphatic
The plausible transition states for the syn-selective aldol
reaction of unprotected hydroxyacetone I catalyzed by 3 is
compared with the transition state of the anti-selective aldol
reaction of 2,2-dimethyl-1,3-dioxan-5-one II catalyzed by l-
proline in Scheme 1. A key feature of the syn-selective
transition state I is the hydrogen-bond-stabilized Z enamine,
which is inaccessible for the reaction of the cyclic ketone 2,2-
dimethyl-1,3-dioxan-5-one under proline catalysis (transition
state II).
Angew. Chem. Int. Ed. 2007, 46, 5572 –5575
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5573

Communications
Table 4: Syn-selective dihydroxyacetone aldol reactions catalyzed by 3.[a]
1,6-diphosphate aldolase, respectively. Thus, the activities of
each of natureꢀs four dihyroxyacetone aldolases can now be
effectively mimicked using organocatalysis. The data pre-
sented here provide further support for our original hypoth-
esis that amino acid catalysis played a key role in prebiotic
chemistry by facilitating the asymmetric synthesis of the
molecules of life.[6a] Further studies concerned with the
expansion of the scope of this chemistry in aldol, Mannich,
and Michael-type reactions are currently underway.
Entry
1
R
Prod. t [days] Yield[b] [%] d.r.[c]
ee [%][c]
(syn/anti) (syn/anti)
8
0.7
76
15:1
15:1
92/20
98/24
Received: March 22, 2007
Published online: June 19, 2007
2.3
2.3
92
2
3
4
5
10
11
12
13
88[d]
>100:1[e] >99[e]
1.9
1.9
82
8:1
10:1
96/74
–
Keywords: aldehydes · aldol reaction · asymmetric catalysis ·
.
91[d]
enamines · organocatalysis
2.1
0.7
85
62
11:1
7:1
92/54
92/12
[1] a) T. M. Wrodnigg, F. K. Sprenger, Mini-Rev. Med. Chem. 2004,
4, 437 – 459; b) D. B. Werz, P. H. Seeberger, Chem. Eur. J. 2005,
11, 3194 – 3206; c) J. J. Barchi, Jr., Curr. Pharm. Des. 2000, 6,
485 – 501.
14
6
2
65
12:1
6
97g]
15[f]
79[g]
33:1[g]
7
8
16
17
3.5
72
7:1
92/62
[2] a) K. C. Nicolaou, H. J. Mitchell, Angew. Chem. 2001, 113, 1624 –
1672; Angew. Chem. Int. Ed. 2001, 40, 1576 – 1624; b) P. H.
Seeberger, D. B. Werz, Nat. Rev. Drug Discovery 2005, 4, 751 –
763; c) P. Sears, C.-H. Wong, Science 2001, 291, 2344 – 2350; d) S.
Hanessian in Preparative Carbohydrate Chemistry, Marcel
Dekker, New York, 1997; e) T. Hudlicky, D. A. Entwistle,
K. K. Pitzer, A. J. Thorpe, Chem. Rev. 1996, 96, 1195 – 1220;
f) S. J. Danishefsky, K. F. McClure, J. T. Randolph, R. B. Rug-
geri, Science 1993, 260, 1307 – 1309; g) S. Y. Ko, A. W. M. Lee, S.
Masamune, L.A. Reed III, K. B. Sharpless, F. J. Walker, Science
1983, 220, 949 – 951.
[3] a) C.-H. Wong, M. C. Bryan, P. T. Nyffeler, H. Liu, E. Chapman,
Pure Appl. Chem. 2003, 75, 179 – 186; b) K. M. Koeller, C.-H.
Wong, Nature 2001, 409, 232 – 240; c) C.-H. Wong, T. D. Macha-
jewski, Angew. Chem. 2000, 112, 1406 – 1430; Angew. Chem. Int.
Ed. 2000, 39, 1352– 1375; d) W. D. Fessner, V. Helaine, Curr.
Opin. Biotechnol. 2001, 12, 574 – 586; e) D. Franke, T. Macha-
jewski, C. C. Hsu, C.-H. Wong, J. Org. Chem. 2003, 68, 6828 –
6831.
3
3
21
28
5:1
9:1
99/28
>99/92[h]
[a] See the Supporting Information for detailed reaction conditions.
Typical reaction conditions: a mixture of aldehyde (0.5 mmol), ketone
(0.5 mmol as dimer, 1 mmol as monomer), catalyst (20 mol%), and 9
(10 mol%) in DMF (0.5 mL) was stirred at RT. [b] Yield of isolated
product after purification by column chromatography. [c] Determined by
HPLC on a chiral stationary phase. [d] Yield of isolated product when the
reaction was performed on 15 mmol scale. [e] After a single recrystal-
lization. [f] Excess aldehyde used. [g] Yield of unacetylated trihydroxyke-
tone 15 after recrystallization; d.r. and ee values calculated after
acetylation. [h] Added 3 vol% of water instead of 9.
[4] W. Notz, F. Tanaka, C. F. Barbas III, Acc. Chem. Res. 2004, 37,
580.
[5] For reviews, see: a) M. Limbach, Chem. Biodiversity 2005, 2,
825 – 836; b) U. Kazmaier, Angew. Chem. 2005, 117, 2224 – 2226;
Angew. Chem. Int. Ed. 2005, 44, 2186 – 2188; c) F. Tanaka, C. F.
Barbas III in Enantioselective Organocatalysis, Reactions and
Experimental Procedures (Ed.: P. I. Dalko), Wiley-VCH, Wein-
heim, 2007, pp. 19 – 55.
[6] For selected references, see: a) N. S. Chowdari, D. B. Ramach-
ary, A. Cordova, C. F. Barbas III, Tetrahedron Lett. 2002, 43,
9591; b) A. B. Northrup, D. W. C. MacMillan, Science 2004, 305,
1752– 1755; c) D. Enders, C. Grondal, Angew. Chem. 2005, 117,
1235 – 1238; Angew. Chem. Int. Ed. 2005, 44, 1210 – 1212; d) J. T.
Suri, D. B. Ramachary, C. F. Barbas III, Org. Lett. 2005, 7, 1383 –
1385; e) D. Enders, C. Grondal, M. Vrettou, G. Raabe ; Angew.
Chem. 2005, 117 4147 – 4151; Angew. Chem. Int. Ed. 2005, 44,
4079 – 4083; f) J. T. Suri, S. Mitsumori, K. Albertshofer, F.
Tanaka, C. F. Barbas III, J. Org. Chem. 2006, 71, 3822 – 3828;
g) D. Enders, M. Vrettou, Synthesis 2006, 2155 – 2158; h) C.
Grondal, D. Enders, Tetrahedron 2006, 62, 329 – 337; i) D.
Enders, J. Palecek, C. Grondal, Chem. Commun. 2006, 655 –
657; j) D. Enders, T. Gasperi, Chem. Commun. 2007, 88 – 90;
k) C. Grondal, D. Enders, Adv. Synth. Catal. 2007, 349, 694 – 702.
Scheme 1. Predicted transition states I and II for the reactions
catalyzed by O-tBu-l-Thr and l-proline, respectively.
In summary, we have developed highly enantioselective
syn-aldol reactions that involve unprotected dihydroxyace-
tone. The unprotected dihydroxyacetone is a significantly
more economical starting material than the protected variants
that have been used in proline- and enzyme-catalyzed
reactions.[10] The reactions catalyzed by O-tBu-l-Thr and O-
tBu-d-Thr are organocatalytic mimics of the reactions cata-
lyzed by l-rhamnulose 1-phosphate aldolase and d-fructose
5574
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5572 –5575

Angewandte
Chemie
[7] For a suggested terminology, see: D. Enders, M. Voith, A.
Lenzen, Angew. Chem. 2005, 117, 1330 – 1351; Angew. Chem. Int.
Ed. 2005, 44, 1304 – 1325.
3R, 4S products prepared here, and it has been stated that the
free hydroxyacetone is not a reactive substrate, see: X. Wu, Z.
Jiang, H.-M. Shen, Y. Lu, Adv. Synth. Catal. 2007, 349, 812– 816.
[9] a) A. Córdova, W. Notz, C. F. Barbas III, Chem. Commun. 2002,
3024 – 3025; b) A. Córdova, W. Zou, P. Dziedzic, I. Ibrahem, E.
Reyes, Y. Xu, Chem. Eur. J. 2006, 12, 5175 and A. Córdova, W.
Zou, P. Dziedzic, I. Ibrahem, E. Reyes, Y. Xu, Chem. Eur. J.
2006, 12, 5383 – 5397.
[10] The price per gram in US dollars is $0.39, $93.00, and $4320.00
for the 1,3-dihydroxyacetone dimer, 2,2-dimethyl-1,3-dioxan-5-
one, and dihydroxyacetone phosphate, respectively.
[8] a) S. S. V. Ramasastry, H. Zhang, F. Tanaka, C. F. Barbas III, J.
Am. Chem. Soc. 2007, 129, 288 – 289; for other syn-selective
organocatalytic aldol reactions that followed, see: b) T. Kano, Y.
Yamaguchi, Y. Tanaka, K. Maruoka, Angew. Chem. 2007, 119,
1768 – 1770; Angew. Chem. Int. Ed. 2007, 46, 1738 – 1740; c) S.
Luo, H. Xu, J. Li, L. Zhang, J.-P. Cheng, J. Am. Chem. Soc. 2007,
129, 3074 – 3075; d) surprisingly, 3S, 4R aldol products have been
reported by using l-threonine-based catalysts, compared to the
Angew. Chem. Int. Ed. 2007, 46, 5572 –5575
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5575