Asymmetric self- and cross-aldol reactions of glycolaldehyde catalyzed by d-fructose-6-phosphate aldolase

Full text of DOI:10.1002/anie.200902065

Angewandte
Chemie
DOI: 10.1002/anie.200902065
Biocatalysis
Asymmetric Self- and Cross-Aldol Reactions of Glycolaldehyde
Catalyzed by d-Fructose-6-phosphate Aldolase**
Xavier Garrabou, Josꢀ A. Castillo, Christine Guꢀrard-Hꢀlaine, Teodor Parella, Jesffls Joglar,
Marielle Lemaire, and Pere Clapꢀs*
Aldol additions are key chemical reactions for the construc-
tion of chiral complex polyhydroxylated molecules.[1–4]
Recent developments in direct aldol additions using bio-,
organo-, and metal catalysts are promising since these
methodologies do not require separate generation of enolate
equivalents and thus improve the atom economy of the
transformation.[1,2,5–8] Aldehydes have been regarded as
highly interesting donors in aldol reactions, because the
products formed are themselves aldehydes that can be used in
further aldol additions for the construction of complex
polyfunctional molecular frameworks.[4] Hence, the direct
catalytic cross-aldol reaction of aldehydes constitutes a
challenge for these methodologies.[4,9,10] Self- and cross-aldol
reactions were achieved by organocatalysis in N,N-dimethyl-
formamide (DMF) using simple aliphatic and aromatic
aldehydes.[11–14] Self- and cross-aldol additions involving
glycolaldehyde derivatives are of paramount interest because
they allow access to polyol architectures.[9,13] Organocatalytic
self- and cross-aldol additions of free glycolaldehyde failed to
provide promising results.[15,16] A successful self-aldol addition
was accomplished in DMF, but it was limited to glycolalde-
hyde derivatives with electron-rich a-alkyloxy or bulky a-
silyloxy protecting groups.[13] No further additions were
observed on the corresponding aldol adducts, a feature
essential for a two-step aldol-based synthesis of carbohy-
drates.[13] This approach was used to prepare protected
hexoses: a direct organocatalytic self-aldol addition was
followed by a direct metal-catalyzed aldol addition.[17] In
cross-aldol additions, the organocatalyst cannot selectively
control the donor and acceptor roles; this is governed by the
aldehyde structure and reactivity.[9] Therefore, in the presence
of simple aliphatic aldehyde donors[9,13] O-protected glycol-
aldehyde derivatives act invariably as acceptors, likely
because they are kinetically disfavored as donors.[18]
Biocatalytic synthetic strategies for carbohydrates and
their analogues require water-soluble polyhydroxyaldehyde
derivatives as acceptor substrates for aldolases.[19,20] Multistep
strategies have suffered from the laborious and costly
isolation of sensitive deprotected hydroxyaldehydes which
are usually obtained by chemical means.[21] In addition, the
vast majority of reported biocatalytically prepared carbohy-
drates and related products are ketoses. This is because
aldolases specific for aldose-type sugars are scarce in nature;
2-deoxyribose-5-phosphate aldolase (DERA) is a notable
exception and actually functions as a deoxysugar aldo-
lase.[22–26] Hence, cross-aldol reactions of aldehydes have
been a limited field for biocatalysis, and DERA is the only
enzyme known to catalyze the stereoselective cross-aldol
addition of acetaldehyde to other aldehydes. However, the
low conversion rates of this enzyme with non-phosphorylated,
unnatural substrates and its inability to generate two consec-
utive hydroxylated positions with each newly formed bond
limit considerably its scope of applicability. Consequently, the
biocatalytic self- and cross-aldol additions of glycolaldehyde
are a challenge for the cascade two-step synthesis of
carbohydrates.
Recently, we reported the synthesis of iminosugars and
other polyhydroxylated compounds catalyzed by d-fructose-
6-phosphate aldolase (FSA).[27,28] This aldolase shows an
unprecedented tolerance for donor substrates such as dihy-
droxyacetone (DHA), hydroxyacetone (HA), and 1-hydroxy-
2-butanone.[28–30] In the course of our investigations on the
catalytic properties of FSA, we discovered a new and
unexpected activity of paramount importance: its ability to
catalyze the direct stereoselective self-aldol addition of
glycolaldehyde (GA) (1) to furnish d-(À)-threose (2)
(Scheme 1).[31] In this reaction, GA (1) acts as both the
[*] X. Garrabou, Dr. J. Joglar, Dr. P. Clapꢀs
Biotransformation and Bioactive Molecules Group
Instituto de Quꢁmica Avanzada de Cataluꢂa-CSIC
Jordi Girona 18-26, 08034 Barcelona (Spain)
Fax: (+34)93-204-5904
E-mail: pere.clapes@iqac.csic.es
Dr. J. A. Castillo, Dr. C. Guꢀrard-Hꢀlaine, Prof. M. Lemaire
Universitꢀ Blaise Pascal, Laboratoire SEESIB-CNRS
UMR 6504-Synthꢃse et Etude de Systꢃmes ꢄ IntꢀrÞt Biologique
24 avenue des Landais, 63177 Aubiꢃre, Aubiꢃre Cedex (France)
Dr. T. Parella
Servei de Ressonꢄncia Magnꢃtica Nuclear
Universitat Autꢅnoma de Barcelona, Bellaterra (Spain)
[**] This work was supported by the Spanish MCINN (CTQ2006-01345/
BQU, CTQ2006-01080), La Maratꢆ de TV3 foundation (Ref:
050931), Generalitat de Catalunya (DURSI 2005-SGR-00698), and
ESF (project COST CM0701). X.G. acknowledges the I3P-CSIC
predoctoral scholarship program. J.A.C. acknowledges the French
foundation Vaincre Les Maladies Lysosomales for a postdoctoral
fellowship. We thank Prof. Wolf-Dieter Fessner, Prof. Georg A.
Sprenger, and Dr. Thierry Gefflaut for their helpful advice and fruitful
discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200902065.
Scheme 1. FSA-catalyzed self-aldol addition of glycolaldehyde (1).
Angew. Chem. Int. Ed. 2009, 48, 5521 –5525
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5521

Communications
donor and acceptor substrate for FSA. This tolerance for an
determined our own parameters for this study (Table 1,
entries 1 and 2). The kinetic model for an enzyme-catalyzed
dimerization reaction developed by Neuhaus[33] was applied
to determine the KM of GA as the donor (entry 4) as well as
the acceptor (Table 1, entry 5) in the self-aldol reaction.
GA and DHA are similar donors in terms of their Vmax
values but GA binds to FSA with higher affinity than DHA,
as indicated by its KM value. Hence, the fact that GA is a
better donor substrate than DHA may explain why when
DHA and GA are mixed in the presence of FSA, 2 is the
major product whereas the expected d-xylulose was not
detected.[28] More importantly, a second in situ cross-aldol
addition of HA to 2 was indeed possible, furnishing 1-deoxy-
d-ido-hept-2-ulose (5)[28] in 68% yield (Scheme 3), whereas
no reaction was detected with either DHA or GA. As judged
by the ratio Vmax/KM, FSA functions with HA approximately
45 times better than with DHA, which might reflect the
different reactivities of these substrates towards aldehyde
acceptors.[28–30] Moreover, 2 binds to FSA at both the acceptor
and donor sites, with an affinity similar to that of GA as the
donor (Table 1, entries 4 and 7); this likely interferes with the
trimerization reaction. The formation of 5 illustrates the
application of a cascade of two consecutive aldol additions
with two different donors catalyzed by FSA. The kinetic
parameters for the FSA-catalyzed retro-aldol reactions of
some sugar derivatives (Table 1, entries 6–8) indicate that the
affinity follows the order d-threose > d-Ara-5-P > d-F6P,
aldehyde was unprecedented and unexpected for FSA and
other related ketose-dependent aldolases (e.g. DHAP aldo-
lases).
We next explored the ability of FSA to catalyze the direct
cross-aldol addition of 1 to other aldehyde acceptors. To this
end, aldehyde 3, a well-known acceptor substrate for FSA,[27]
was selected as an example. Interestingly, a preliminary
experiment gave 70% conversion of 3 to the cross-aldol
adduct 4 (> 97% syn)[32] as a mixture of the two cyclic
isomeric forms 4a and 4b (Scheme 2; see the Supporting
Information).
Scheme 2. Cross-aldol addition of 1 to 3 catalyzed by FSA. Cbz=ben-
zyloxycarbonyl.
To understand and further exploit this new catalytic
potential of FSA for the direct cross-aldol
additions of aldehydes, we determined the
kinetic parameters for enzymatic aldol reac-
tions with dihydroxyacetone (DHA) and
hydroxyacetone (HA) as donors and GA as
both donor and acceptor (Table 1). In addi-
tion, the kinetic parameters for the retro-
aldol reaction with d-fructose-6-phosphate
(d-F6P), d-arabinose-5-phosphate (d-Ara-5-
P), and d-threose were measured. As the
reported kinetic parameters for DHA and
Scheme 3. FSA-catalyzed dimerization of GA (1) and a second aldol addition of HA to
HA have significant discrepancies,[29,30] we d-(À)-threose in situ.
and, as Vmax/KM reflects, d-Ara-5-P is the best substrate in
this series. Hence, the metabolic function of FSA, which
remains uncertain, may not be restricted to the cleavage of d-
F6P; FSA might have a different, even multipurpose function
in carbohydrate metabolism that now warrants further
investigation.
The kinetic data (Table 1) indicate that the apparent
catalytic efficiency for GA is approximately 320-fold better as
the donor than as the acceptor. This is of paramount
importance since it makes possible to optimize the cross-
aldol addition by controlling the concentration of GA during
the reaction. We examined in detail the syn-selective cross-
aldol addition of 1 to 3. We determined the initial rates of
cross-aldol (vo) and self-aldol (i.e. formation of 2) (vo’)
reactions at different concentrations of [1] and [3] to shed
light on the reaction preferences (Table 2). In good agree-
ment with the KM value, when [1] < 50 mm and with an excess
of 3, vo > vo’, and the cross-aldol reaction is favored (Table 2).
Table 1: Steady-state kinetic parameters of aldol and retro-aldol
reactions catalyzed by d-fructose-6-phosphate aldolase.
Entry Compound
K
M [mm]
Vmax [U[e] mgÀ1
powder]
(Vmax/KM)/10À6
[minÀ1 mgÀ1
]
1
2
3
4
5
6
7
8
DHA[a]
32Æ2
1.46Æ0.05
33.7Æ0.9
1.5Æ0.3
}0.22Æ0.02
0.34Æ0.02
46Æ3
1937Æ76
nd[c]
1117Æ274
3.5Æ0.4
37
HA[a]
17.4Æ0.5
nd[c]
GA (1)[a]
GA(don.)[b]
GA(acc.)[b]
d-F6P
0.197Æ0.045
62.8Æ5.5
9[d]
d-threose (2) 0.286Æ0.017 (1.15Æ0.04)ꢇ10À3 4.0Æ0.3
d-Ara-5-P
0.561Æ0.042 0.160Æ0.012
285Æ30
[a] Aldol addition to d,l-glyceraldehyde-3-phosphate (d,l-G3P). [b] Self-
aldol addition of GA. [c] nd: not determined; the KM value of GA, as
donor, for the reaction GA+d,l-G3P could not be correctly measured,
because of the competing self-aldol addition of GA. [d] Data from
Ref. [34]. [e] U: mmol consumed per minute. Specific activity of d-
fructose-6-phosphate (d-F6P) cleavage=0.34Æ0.02 UmgÀ1 FSA
powder. (see the Supporting Information for further details)
5522
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 5521 –5525

Angewandte
Chemie
Table 2: FSA-catalyzed cross-aldol additions of GA (1) to 3. Initial
reaction rates of the cross-aldol addition (vo) and the competing self-
aldol addition (vo’) of 1.
[1], [3]
[mm, mm]
vo cross-aldol
vo’ auto-aldol
vo/vo’
[mmolminÀ1
]
[mmolminÀ1
]
10, 100
25, 100
50, 100
25, 25
50, 50
100, 50
0.37
0.15
0.09
0.07
0.04
0.03
0.04
0.06
0.11
0.06
0.11
0.12
9.3
2.3
0.8
1.1
0.4
0.2
On the other hand, the higher the concentration of 1, the
faster the rate vo’ achieved regardless of the concentration of
3. In practice, it is convenient to keep [1] < 5 mm in the
reaction mixture to ensure a saturation concentration of 1 as
the donor and well under saturation levels as the acceptor,
therefore resulting in a high ratio of cross- to self-aldol
reaction rates. We conducted a set of experiments (see the
Supporting Information) in which a solution of 1 was added
with a syringe pump in order to approximate a constant
concentration of 1; the rate of the cross-aldol reaction was
found to be much higher than that of the self-aldol (Figure 1).
Moreover, the d-(À)-threose generated reacted further, at
much a slower rate (as expected according to its Vmax/KM) to
furnish 4a and 4b under equilibrium conditions. These
conditions can be generalized for other acceptor aldehydes;
however, the outcome will depend upon the substrate quality
(i.e. the KM value should be lower than that of 1 as the
acceptor) and the tendency for cyclization of the adduct
formed (e.g. 4a and 4b).
Figure 1. Progress of the FSA-catalyzed direct aldol addition of GA (1)
to 3. a) Initial amount of reactants [1]=[3]=25 mm; amounts of
~
&
^
cross-aldol product 4 ( ), self-aldol product 2 ( ), and 1 ( ) in
solution. b) Total 1 added 2.2 mmol, 3=2 mmol; amounts of cross-
&
^
aldol product 4 ( ), self-aldol product 2 ( ), 1 added by a syringe
*
pump ( ). See the Supporting Information for details.
In further examples (Scheme 4) the FSA catalyst showed
exquisite syn stereoselectivity and good to excellent con-
versions, and provided the products in yields between 28 and
68% under nonoptimized reaction conditions and purifica-
tion procedures. l-Lactaldehyde (6), easily obtained by a
previously described methodology,[35,36] furnished 5-deoxy-l-
xylose (7) in a single step. In this case, vo/vo’ = 1.24 at a
concentration of 25 mm for each substrate, indicating that 6 is
a relatively good acceptor. Furthermore, we observed that
when a mixture of 2 and 6 (2 equiv) was treated with FSA, 7
was obtained, indicating that both equilibrium thermody-
namics and kinetics favor 7 and that 2 acts as donor substrate
as well (i.e. by means of a retro-aldol/aldol sequence).
Racemic 8 was also accepted yielding the iminosugar
cyclic derivatives 9a and 9b as major products; the non-
cyclized isomer was not detected by NMR spectroscopy. In
this case, quantitative conversion was achieved in the first
tests indicating that 8 is a good acceptor (see the Supporting
Information). Removal of the Cbz protecting group of 9
Scheme 4. FSA-catalyzed aldol additions of 1 and selected aldehydes.
Conversions of the enzyme-catalyzed reaction are given under the reaction
arrows.
Angew. Chem. Int. Ed. 2009, 48, 5521 –5525
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5523

Communications
[7] C. Bolm, S. Brꢀse, H.-J. Gais, G. Franciꢁ, F. Faraone, W. Leitner,
A. Crosman, C. Schuster, H.-H. Wagner, M. Batorfi, J. Cubillos,
W. Hꢂlderich in Asymmetric Synthesis with Chemical and
Biological Methods (Eds.: D. Enders, K.-E. Jaeger), Wiley-
VCH, Weinheim, 2007, pp. 149 – 297..
[8] C. Palomo, M. Oiarbide, J. M. Garcꢃa, Chem. Soc. Rev. 2004, 33,
65 – 75.
[9] G. Guillena, C. Nꢄjera, D. J. Ramꢅn, Tetrahedron: Asymmetry
2007, 18, 2249 – 2293.
[10] L. M. Geary, P. G. Hultin, Tetrahedron: Asymmetry 2009, 20,
131 – 173.
[11] A. B. Northrup, D. W. C. MacMillan, J. Am. Chem. Soc. 2002,
124, 6798 – 6799.
[12] H. Liu, L. Peng, T. Zhang, Y. Li, New J. Chem. 2003, 27, 1159 –
1160.
[13] A. B. Northrup, I. K. Mangion, F. Hettche, D. W. C. MacMillan,
Angew. Chem. 2004, 116, 2204 – 2206; Angew. Chem. Int. Ed.
2004, 43, 2152 – 2154.
followed by reductive amination led to the 1-deoxyiminosu-
gars 10a and 10b, which are of interest as glycosidase
inhibitors.[37–39] 3-Benzyloxy-d,l-glyceraldehyde (11) was a
good substrate and furnished a mixture of 5-O-benzyl-
protected precursors of l-xylose and d-arabinose (12a
(67%) and 12b (33%), respectively). Reaction with formal-
dehyde (13) furnished l-glyceraldehyde (14) quantitatively;
vo/vo’ was 2.22 at substrate concentrations of 25 mm, as
expected since 13 is an excellent substrate in enzymatic aldol
additions.[40,41] Phenylacetaldehyde (15) was also a substrate
with a good vo/vo’ ratio (i.e. 1.39 at [15] = [1] = 25 mm).
Interestingly, a single aldol addition was observed in all the
examples under the reaction conditions employed; this is also
of great value for a two-step biocatalytic aldol addition for the
stereoselective synthesis of carbohydrates.
In conclusion, we have discovered a new general activity
of FSA, unprecedented for ketose-dependent aldolases. As a
result, a new biocatalytic strategy based on FSA for the direct
syn cross-aldol addition of glycolaldehyde to aldehydes in
aqueous media has been developed. As the high affinity of
GA for FSA as a donor is higher than that as an acceptor,
cross-aldol reactions are possible with good to poor aldehyde
acceptors if the concentration of GA in the reaction is kept
low. Significantly, the control of the nucleophile and the
electrophilic acceptor depend on FSA regardless of the
absolute configuration and the intrinsic relative reactivity of
the aldehydes, in contrast to what occurs in organocatalytic
reactions. Hence, in the cross-aldol additions of GA to the
aldehydes selected, the former acted invariably as the donor.
The fact that aldol adducts are inert to further GA additions is
advantageous. Therefore, new synthetic strategies can be
easily devised through in situ or enzyme-catalyzed two-step
aldol cascade reactions. In fact, by FSA catalysis we have
efficiently generated various aldehydes that were previously
used in reported enzymatic aldol reactions.[21,42–45] Structure-
based site-directed mutagenesis of the FSA active center and
exploitation of directed evolution to modify its donor or
acceptor substrate tolerance and the stereochemical outcome
of the reaction can expand the synthetic possibilities. Work in
this direction is currently in progress in our lab.
[14] T. Kano, Y. Yamaguchi, Y. Tanaka, K. Maruoka, Angew. Chem.
2007, 119, 1768 – 1770; Angew. Chem. Int. Ed. 2007, 46, 1738 –
1740.
[15] S. Pizzarello, A. L. Weber, Science 2004, 303, 1151.
[16] J. Kofoed, M. Machuqueiro, J. L. Reymond, T. Darbre, Chem.
Commun. 2004, 1540 – 1541.
[17] A. B. Northrup, D. W. C. MacMillan, Science 2004, 305, 1752 –
1755.
[18] The control of the donor and acceptor aldehyde depended upon
their intrinsic reactivity: the donors always possess an accessible
enolizable a proton (i.e. kinetically favored), whereas the
acceptors are sterically hindered, b,b-disubstituted, or kinetically
inaccessible a,a-disubtituted aldehydes.
[19] W.-D. Fessner, C. Walter, Top. Curr. Chem. 1997, 184, 97 – 194.
[20] W.-D. Fessner, V. Helaine, Curr. Opin. Biotechnol. 2001, 12, 574 –
586.
[21] I. Henderson, K. B. Sharpless, C.-H. Wong, J. Am. Chem. Soc.
1994, 116, 558 – 561.
[22] C. F. Barbas III, Y. F. Wang, C.-H. Wong, J. Am. Chem. Soc.
1990, 112, 2013 – 2014.
[23] C.-H. Wong, E. Garcia-Junceda, L. Chen, O. Blanco, H. J. M.
Gijsen, D. H. Steensma, J. Am. Chem. Soc. 1995, 117, 3333 –
3339.
[24] L. Chen, D. P. Dumas, C. H. Wong, J. Am. Chem. Soc. 1992, 114,
741 – 748.
[25] H. J. M. Gijsen, C.-H. Wong, J. Am. Chem. Soc. 1994, 116, 8422 –
8423.
[26] H. J. M. Gijsen, C.-H. Wong, J. Am. Chem. Soc. 1995, 117, 7585 –
7591.
Received: April 17, 2009
Published online: June 24, 2009
[27] J. A. Castillo, J. Calveras, J. Casas, M. Mitjans, M. P. Vinardell, T.
Parella, T. Inoue, G. A. Sprenger, J. Joglar, P. Clapꢆs, Org. Lett.
2006, 8, 6067 – 6070.
[28] A. L. Concia, C. Lozano, J. A. Castillo, T. Parella, J. Joglar, P.
Clapꢆs, Chem. Eur. J. 2009, 15, 3808 – 3816.
Keywords: aldol reaction · aldolases · aldoses · enzyme catalysis ·
.
glycolaldehydes
[29] M. Schꢇrmann, M. Schꢇrmann, G. A. Sprenger, J. Mol. Catal. B
2002, 19, 247 – 252.
[30] M. Sugiyama, Z. Hong, P. H. Liang, S. M. Dean, L. J. Whalen,
W. A. Greenberg, C.-H. Wong, J. Am. Chem. Soc. 2007, 129,
14811 – 14817.
[1] R. Mahrwald, Modern Aldol Reactions, Vol. 1, Wiley-VCH,
Weinheim, 2004.
[31] Within the limits of detection by high-field 1H and 13C NMR
spectroscopy a single diasteromer was observed; see the in situ
NMR spectra recorded during the reaction and the NMR
spectrum of an authentic sample of d-(À)-threose in the
Supporting Information. The value of the optical rotation is in
agreement with published values for d-(À)-threose. The stereo-
selectivity observed was thus (2S,3R)-threo in agreement with
previous work with other FSA-catalyzed aldol additions.[27–30]
[32] Within the detection limits of high-field 1H and 13C NMR
spectroscopy, only the syn diasteromer was observed.
[2] R. Mahrwald, Modern Aldol Reactions, Vol. 2, Wiley-VCH,
Weinheim, 2004.
[3] D. A. Evans, J. V. Nelson, T. R. Taber, Top. Stereochem. 1982, 13,
1 – 115.
[4] S. Mukherjee, J. W. Yang, S. Hoffmann, B. List, Chem. Rev. 2007,
107, 5471 – 5569.
[5] D. Enders, A. A. Narine, J. Org. Chem. 2008, 73, 7857 – 7870.
[6] W.-D. Fessner in Asymmetric Synthesis with Chemical and
Biological Methods (Eds.: D. Enders, K.-E. Jaeger), Wiley-
VCH, Weinheim, 2007, pp. 351 – 375.
5524
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 5521 –5525

Angewandte
Chemie
[33] F. C. Neuhaus, J. Biol. Chem. 1962, 237, 3128 – 3135.
[34] M. Schꢇrmann, G. A. Sprenger, J. Biol. Chem. 2001, 276, 11055 –
11061.
[35] W. D. Fessner, A. Schneider, O. Eyrisch, G. Sinerius, J. Badia,
Tetrahedron: Asymmetry 1993, 4, 1183 – 1192.
[41] W.-D. Fessner, G. Sinerius, A. Schneider, M. Dreyer, G. E.
Schulz, J. Badia, J. Aguilar, Angew. Chem. 1991, 103, 596 – 599;
Angew. Chem. Int. Ed. Engl. 1991, 30, 555 – 558.
[42] R. Alajarꢃn, E. Garcꢃa-Junceda, C.-H. Wong, J. Org. Chem. 1995,
60, 4294 – 4295.
[36] B. Zagalak, P. A. Frey, G. L. Karabatsos, R. H. Abeles, J. Biol.
Chem. 1966, 241, 3028 – 3035.
[37] G. Pandey, K. C. Bharadwaj, M. I. Khan, K. S. Shashidhara, V. G.
Puranik, Org. Biomol. Chem. 2008, 6, 2587 – 2595.
[38] R. C. Bernotas, G. Papandreou, J. Urbach, B. Oanem, Tetrahe-
dron Lett. 1990, 31, 3393 – 3396.
[39] T. Sekioka, M. Shibano, G. Kusano, Nat. Med. 1995, 49, 332 – 335.
[40] M. D. Bednarski, E. S. Simon, N. Bischofberger, W.-D. Fessner,
M. J. Kim, W. Lees, T. Saito, H. Waldmann, G. M. Whitesides, J.
Am. Chem. Soc. 1989, 111, 627 – 635.
[43] M. D. Bednarski, D. C. Crans, R. DiCosimo, E. S. Simon, P. D.
Stein, G. M. Whitesides, M. J. Schneider, Tetrahedron Lett. 1988,
29, 427 – 430.
[44] T. Sugai, G. J. Shen, Y. Ichikawa, C. H. Wong, J. Am. Chem. Soc.
1993, 115, 413 – 421.
[45] W. Fitz, J.-R. Schwark, C.-H. Wong, J. Org. Chem. 1995, 60,
3663 – 3670.
Angew. Chem. Int. Ed. 2009, 48, 5521 –5525
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5525