Highly efficient and versatile synthesis of proteoglycan core structures from 1,6-anhydro-β-lactose as a key starting material

Article abstract of DOI:10.1002/anie.200604909

Versatile building blocks in the form of novel derivatives of 1,6-anhydro-β-lactose have been used for the rapid and highly efficient synthesis of various proteoglycan core structures (see example). The resulting oligosaccharide-tetrapeptide conjugates ca

Full text of DOI:10.1002/anie.200604909

Communications
DOI: 10.1002/anie.200604909
Oligosaccharide Synthesis
Highly Efficient and Versatile Synthesis of Proteoglycan Core
Structures from 1,6-Anhydro-b-lactose as a Key Starting Material**
Ken Shimawaki, Yoshinori Fujisawa, Fumihiro Sato, Naoki Fujitani, Masaki Kurogochi,
Hiroko Hoshi, Hiroshi Hinou, and Shin-Ichiro Nishimura*
Cell-surface and extracellular-matrix glycosaminoglycans
(GAGs) of proteoglycans (PGs) are known to play vital
functional roles in various biological processes, such as cell
proliferation, differentiation, adhesion, blood coagulation,
and wound repair.[1] The major GAG chains found in PGs are
divided into two categories, heparan sulfate/heparin and
chondroitin sulfate/dermatan sulfate, which are high-molec-
ular-weight polysaccharides that consist of repeating
[GlcNAca(1!4)GlcAb(1!4)] and [GalNAcb(1!4)GlcAb-
(1!3)] units, respectively.These GAG chains are further
modified by some epimerases and sulfotransferases to pro-
vide considerable complexity and the functionality required
for interaction with growth factors, cytokines, and many
protein ligands.The biosynthetic pathway to the mature GAG
chains is initiated from the core tetrasaccharide moiety,
namely, GlcAb(1!3)Galb(1!3)Galb(1!4)Xylb1!, which
is attached to the serine residue of core proteins.Modification
of this GAG-protein linkage region by phosphorylation and/
or sulfation seems to influence the molecular diversity in the
biosynthetic pathway of the disaccharide repeating struc-
tures.[2,3] Furthermore, GAG precursors have been proposed
as antiamyloid or antithrombotic agents in anticancer ther-
apy.[4] Although extensive efforts have been made toward the
synthesis of GAGs, with particular emphasis on the core
structures,[3,5] we believed that an efficient and versatile
strategy for assembling various proteoglycan initiators would
greatly accelerate the process of gaining an understanding of
the functional roles of the enzymes responsible for the
construction of GAGs.We communicate herein a highly
efficient and versatile synthesis of proteoglycan core struc-
tures via key intermediates derived from 1,6-anhydro-b-
lactose,[6] which can be prepared readily from naturally
occurring, abundant lactose (Figure 1).We demonstrate the
[*] K. Shimawaki, Y. Fujisawa, F. Sato, Dr. N. Fujitani, Dr. M. Kurogochi,
Dr. H. Hoshi, Dr. H. Hinou, Prof. Dr. S.-I. Nishimura
Laboratory for Advanced ChemicalBiool gy
Figure 1. A synthetic strategy based on the conversion of 1,6-anhydro-
b-lactose into the key intermediates 1 and 9a for the construction of
proteoglycan core linkage structures. Bn=benzyl.
Graduate Schoolof Advanced Life Science
Frontier Research Center for Post-Genome Science and Technology
Hokkaido University, Sapporo 001-0021 (Japan)
Fax: (+81)11-706-9042
versatility of the building blocks 1 and 9a in the synthesis of
model tetrapeptides H-Ser1-Gly2-Ser3-Gly4-OH with di-
(Galb(1!4)Xylb1!), tri- (Galb(1!3)Galb(1!4)Xylb1!),
E-mail: shin@glyco.sci.hokudai.ac.jp
Dr. H. Hinou, Prof. Dr. S.-I. Nishimura
Drug-Seeds Discovery Research Laboratory
NationalInstitute of Advanced IndustrialScience and Technool gy
(AIST)
Sapporo 062-8517 (Japan)
Fax: (+81)11-857-8451
and
tetrasaccharide
(GlcAb(1!3)Galb(1!3)Galb(1!
4)Xylb1!) branches at the hydroxy group of the Ser3 residue.
As found in pioneering studies on the synthesis of
GAGs,[5a–d] d-xylopyranose often makes selective chemical
modification difficult.It seems likely that regioselective
modifications and the stereoselective glycosidation of d-
xylose derivatives would entail complicated synthetic proto-
cols with tedious multistep protection/deprotection processes
and the purification of anomeric mixtures from glycosidation
reactions.We directed our attention toward reports that 1,6-
anhydro-b-lactose can be used for the efficient synthesis of
various important oligosaccharide derivatives.[6] We hypothe-
sized that the use of 1,6-anhydro-b-lactose would permit the
E-mail: tiger.nishimura@aist.go.jp
[**] This research was partly supported by a grant for the “National
Project on Functional Glycoconjugate Research Aimed at Develop-
ing New Industry” from the Ministry of Education, Science, and
Culture of Japan. We thank S. Oka, M. Kiuchi, and T. Hirose of the
Center of InstrumentalAnaylsis, Hokkaido University, for high
resolution mass spectrometric measurements and amino acid
analysis.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
3074
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 3074 –3079

Angewandte
Chemie
rapid and efficient synthesis
of a variety of Galb(1!
4)Xyl derivatives suitable
for the construction of
more complex and longer
oligosaccharides as ana-
logues of the linkage
region of the proteoglycans.
Scheme 1 shows a syn-
thetic route to the key inter-
mediate 1, which is suitably
functionalized for both
selective elongation with
sugar residues at C3’ and
glycosidation with core pep-
tides at the anomeric (C1)
position, and 9a, which can
also undergo glycosidation
with core peptides at the
anomeric position.The syn-
Scheme 1. Synthetic route to the key intermediate 1: a) NaOMe/MeOH; b) BnBr, NaH, DMF for 3a; 1) nBu2SnO,
toluene, reflux; then allyl bromide, tetra-n-butylammonium bromide; 2) BnBr, NaH, DMF for 3b; c) TMSSPh,
TMSOTf, CH2Cl2; d) 1) Ac2O, pyridine; 2) [Ir(cod)(PMePh2)2PF6], H2, EtOH; then 1n HCl, acetone; 3) TBSOTf,
2,6-lutidine, CH2Cl2; 4) NaOMe/MeOH; e) 1) TsCl, pyridine; 2) NaI, 2-butanone; f) DBU, DMF, reflux; g) O3,
MeOH/THF; h) 1) DIBAL-H, CH2Cl2; 2) (COCl)2, DMF, CH2Cl2; 3) NaBH3CN, THF, reflux for 9a, and LiEt3BH,
THF, reflux for 9b; i) TBAF, THF. cod=cyclooctadiene, DMF=N,N-dimethylformamide, TBS=tert-butyldimethyl-
silyl, Tf=trifluoromethanesulfonyl, Ts=p-toluenesulfonyl.
thesis of compound
1
involves 1) regioselective
modifications at the 3’-posi-
tion of 1,6-anhydro-b-lac-
tose, and 2) ring opening at
the 1,6-anhydro bond and
deletion of the hydroxy-
methyl group of the d-glu-
copyranose residue for
replacement with a d-xylo-
pyranose residue.First, the
known compound 2[6c] was
subjected to de-O-acetyla-
tion and subsequent stanny-
lation[7] with n-dibutyltin
oxide (nBu2SnO), followed
by the introduction of an O-
allyl group at the 3’-position
and perbenzylation, to give
compound 3b in 87% over-
all yield.(Compound
3a
was prepared readily by
direct O benzylation of
free 1,6-anhydro-b-lactose
in 93% yield.) The C6 hy-
droxy group was liberated
by treatment with phenyl-
thio(trimethyl)silane
(TMSSPh) and trimethyl-
silyl trifluoromethanesulfo-
nate (TMSOTf)[8] to afford
alcohols 4a (79%) and 4b
(70%).The allyl protecting
group in 4b was exchanged
for a tert-butyldimethylsilyl
group (5, 72% over four
steps) for the planned ozo-
nolysis.Next, the iodinated
Scheme 2. Synthetic route to the thiophenylgylcosides 11 and 19: a) TMSOTf, MS (AW300), CH2Cl2/CH3CN;
b) 1) DTBS(OTf)2, 2,6-lutidine, CH2Cl2; 2) BzCl, pyridine, DMAP; c) [Ir(cod)(PMePh2)2PF6], H2, THF; then
TsOH·H2O; d) TMSOTf, MS (AW300), CH2Cl2/CH3CN, 15; e) 1) (HF)x·pyridine, THF; 2) Ac2O, pyridine;
f) 1) CAN, CH3CN/H2O; 2) CCl3CN, DBU. Bz=benzoyl, CAN=ceric ammonium nitrate, DTBS=di-tert-
derivatives 6a and 6b were butylsilyl, DMAP=4-dimethylaminopyridine, MP=4-methoxyphenyl, MS=molecular sieves.
Angew. Chem. Int. Ed. 2007, 46, 3074 –3079
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3075

Communications
prepared by tosylation and a subsequent substitution reaction
48% yield, respectively (Scheme 3).Finally, the protecting
groups were removed to afford the target glycopeptides 22
(100%), 24 (78%), and 26 (40%).Precise structural charac-
terization of the proteoglycan linkage region was carried out
for the first time by analysis by mass spectrometry (ESI and
with sodium iodide in 87% and 89% yield, respectively.The
b elimination of 6a and 6b with 1,8-diazabicyclo[5,4,0]undec-
7-ene (DBU) gave the unsaturated exo-methylene derivatives
7a (57%) and 7b (75%), respectively, the treatment of which
with O3 afforded lactones 8a (100%) and 8b (90%).The
hemiacetals generated by the reduction of 8a and 8b in the
presence of diisobutylaluminum hydride (DIBAL-H) were
subjected to chlorination with oxalyl chloride and subsequent
reduction by treatment with sodium cyanoborohydride to
afford 9a in 47% and 9b in 71% yield (over three steps).
Finally, the removal of the silyl protecting group in 9b by
using tetra-n-butylammonium fluoride (TBAF) gave the key
building block 1 in 98% yield.
Scheme 2 shows the synthetic procedures for the con-
struction of the tri- and tetrasaccharide derivatives 11 and 19,
the glycosyl donors later conjugated with the tetrapeptide
derivative 20.Compound 11 was prepared in 58% yield by
the glycosidation of 1 with the known imidate 10 in the
presence of TMSOTf.[9] In the case of compound 19, we had to
establish an improved synthetic pathway for the construction
of the imidate 18, the designated glycosyl donor with a
GlcAb(1!3)Gal moiety.Owing to the extremely low reac-
tivity of compound 15 as a result of the electron-withdrawing
carboxy group, we chose to pre-
1
MALDI) and H,13C NMR spectroscopy.Fragmentation by
MALDI-LIFT-TOF/TOF mass spectrometry[13] of the pre-
cursor ion derived from 26 (m/z 939.39) in the presence of 2,5-
dihydroxybenzoic acid (DHB) occurred both at the glycoside
linkages and at the peptide moiety to give meaningful ion
peaks generated by the characteristic cleavage patterns
represented in Figure 2.The present large-scale synthesis
made a systematic study by high-resolution NMR spectros-
copy possible.HSQC and TOCSY spectra [14] of 26 acquired at
600 MHz allowed complete assignment of 1H and 13C
chemical shifts (Figure 3).
We tested the suitability of compounds 22 (disaccharide–
tetrapeptide: 2s4p), 24 (3s4p), and 26 (4s4p) as proteoglycan
initiators (substrates) for the elongation of GAG chains by
using normal human articular chondrocyte (nHAC-Kn).The
compounds were labeled with fluorescein isothiocyanate
(FITC), then incubated with nHAC-Kn (1.0 ” 105 cells) at
378C for 48 h.It was demonstrated that all synthetic initiators
can be incorporated into cultured cells and modified by the
pare the glycosyl acceptor 14
with a cyclic di-tert-butylsilylene-
diyl ether as a protecting group
for the hydroxy groups at C4 and
C6[10] as reported by Thollas and
Jacquinet.[11] As anticipated, the
enhanced nucleophilicity of the
C3 hydroxy group of 14 due to
the
conformational
stress
induced by this protecting group
facilitated its reaction with 15 to
give the disaccharide derivative
16 in high yield (97%).After the
removal of the silyl group and
acetylation of the resulting diol
(74% over two steps), the
replacement of the methoxy-
phenyl group at the anomeric
position with a trichloroacetimi-
date group afforded the desired
glycosyl donor 18 in a satisfactory
yield of 56% from 17.The cou-
pling reaction between 18 and 1
proceeded smoothly in the pres-
ence of TMSOTf to give the
tetrasaccharide 19 in 66% yield.
The thioglycosides 9a, 11,
and 19 underwent glycosidation
with the tetrapeptide derivative
20 in the presence of trifluoro-
methanesulfonic acid (TfOH)
and N-iodosuccimide (NIS)[12] to
give the fully protected glycopep-
tides 21, 23, and 25 in 74, 34, and
Scheme 3. Coupling of the thioglycosides with the tetrapeptide 20: a) TfOH, NIS, MS (4Š), CH2Cl2/
CH3CN; b) Pd/C, H2 for 22; 1) Pd/C, H2; 2) NaOMe/MeOH for 24; 1) Pd/C, H2; 2) aqueous LiOH;
3) aqueous NaOH/MeOH for 26.
3076 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 3074 –3079

Angewandte
Chemie
Figure 2. MALDI-LIFT-TOF/TOF MS analysis of the proteoglycan core glycopeptide
26; b-17-GlcA-Gal-Gal 3 signifies the b3 fragment ion less 17 (NH3), GlcA, Gal, and
Gal.
biosynthetic pathways of GAG-chain-extension enzymes
(Figure 4).Furthermore, it seems that 22 (2s4p) is a much
better substrate than 24 (3s4p) and 26 (4s4p), thus indicating
that the carbohydrate structure and/or molecular size of the
initiator might strongly influence how well it is accepted by
enzymes or the efficiency of its incorporation by nHAC-Kn.
To verify this hypothesis, we synthesized tandem-repeating
polypeptides from 22 as a macromonomer by direct polycon-
densation reactions according to the method reported pre-
Figure 3. High-resolution NMR spectroscopic analysis of the proteogly-
can core glycopeptide 26: a) 13C-edited HSQC spectrum; b) TOCSY
spectrum.
viously (Figure 5).[15] We could conclude clearly that macro-
molecular initiators from the dimer (MW= 1186.61) to the
heptamer (MW= 4145.60) can be used as good substrates for
cellular GAG biosynthetic processes and afford a product
fraction of high molecular weight estimated at 86 kDa
(Figure 5b).These results indicate that the acceptability of
synthetic glycopeptide initiators by human chondrocytes
depends greatly on the carbohydrate structure in the linker
moiety rather than on the molecular weight.
intermediate, phenyl O-(2,4,6-tri-O-benzyl-b-d-galactosyl)-
(1!4)-2,3-di-O-benzyl-1-thio-a-d-xylopyranoside (1), was
synthesized by selective chemical modifications at the 6-
position of the reducing Glc residue and the 3’-position of the
nonreducing Gal residue in high overall yield.The general
synthetic route established herein facilitates the systematic
and rapid synthesis of tetrapeptides (H-Ser1-Gly2-Ser3-Gly4-
OH) with di-, tri-, and tetrasaccharide branches, such as
Galb(1!4)Xylb1! (in 22), Galb(1!3)Galb(1!4)Xylb1!
(in 24), and GlcAb(1!3)Galb(1!3)Galb(1!4)Xylb1! (in
26), at the hydroxy group of the Ser3 residue.These
derivatives, which are related to proteoglycan core structures,
In conclusion, we have demonstrated the versatility of 1,6-
anhydro-b-lactose in the rapid, efficient, and large-scale
synthesis of important proteoglycan core structures.A key
Angew. Chem. Int. Ed. 2007, 46, 3074 –3079
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3077

Communications
Figure 4. HPLC assay for sugar extension by human chondrocytes with the FITC-labeled initiators 22, 24, and 26. tR =retention time.
Figure 5. Tandem-repeating glycopeptide 28 derived from 22 and acceptability by human chondrocytes: a) Synthesis of tandem-repeating
polypeptides by the diphenylphosphoryl azide (DPPA) method: 1) Ac2O, DMF; 2) DPPA (1.2 equiv), Et3N (2.2 equiv), DMF; b) MALDI-TOF mass
spectrum showing the molecular weights of the polypeptides derived from 22; c) HPLC assay of GAG extension by human chondrocytes; polymer
28 was used after conjugation with FITC at the C terminus.
B.B. Olwin, Science 1991, 252, 1705 – 1708; d) D.M. Ornitz,
A.B.Herr, M.Nilsson, J.Westman, C.M.Svahn, G.Waksman,
Science 1995, 268, 432 – 436; e) C.T.Esmon, J. Biol. Chem. 1989,
264, 4743 – 4746; f) S.Mizuguchi, T.Uyama, H.Kitagawa, K.H.
Nomura, K.Dejima, K.Gengyo-Ando, S.Mitani, K.Sugahara,
K.Nomura, Nature 2003, 423, 443 – 448.
can be used as potent synthetic primers in human cartilage
cells to produce novel proteoglycans with mature glycosami-
noglycan chains of high molecular weight.It is our belief that
the present approach will greatly accelerate the acquisition of
fundamental insight into the functional roles of the essential
core linker moiety of the glycosaminoglycans, and will be able
to be applied to the study of artificial proteoglycans in
therapeutic and biomedical fields.
[2] a) N.R. Krishna, P.K. Agrawal, Adv. Carbohydr. Chem. Bio-
chem. 2000, 56, 201 – 234; b) R.M. Lauder, T.N. Huckerby,
G.M. Brown, M.T. Bayliss, I.A. Nieduszynski,
Biochem. J.
2001, 358, 523 – 528; c) S.Yamada, M.Oyama, Y.Yuki, K.Kato,
K.Sugahara, Eur. J. Biochem. 1995, 233, 687 – 693.
Received: December 5, 2006
Published online: March 15, 2007
[3] a) B-.T.Kim, H.Kitagawa, J.Tanaka, J-.I.Tamura, K.Sugahara,
J. Biol. Chem. 2003, 278, 41618 – 41623; b) S.Gulberti, V.
Lattard, M.Fondeur, J-.C.Jacquinet, G.Mulliert, P.Netter, J.
Magdalou, M.Ouzzine, S.Fournel-Gigleux, J. Biol. Chem. 2004,
280, 1417 – 1425.
[4] a) R.Kisilevasky, W.A.Szarek, J.Ancsin, R.Vohra, Z.Li, S.
Marone, J. Mol. Neurosci. 2004, 4, 167 – 172; b) N.B.Martin, P.
Sepulchre, J.Theveniaux, J.Millet, F.Bellamy, Semin. Thromb.
Hemostasis 1996, 22, 247 – 254; c) M.Belting, L.Borsig, M.M.
Keywords: biosynthesis · carbohydrates · extracellular matrices ·
.
glycosaminoglycans · proteoglycans
[1] a) L.KjellØn, U.Lindahl, Annu. Rev. Biochem. 1991, 60, 443 –
475; b) A.Yayon, M.Klagsbrun, J.D.Esko, P.Leder, D.M.
Ornitz, Cell 1991, 64, 841 – 848; c) A.C.Rapraeger, A.Krufka,
Fuster, J.R.Brown, L.Persson, L.A.Fronsson, J.D.Esko,
Proc.
3078
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 3074 –3079

Angewandte
Chemie
Natl. Acad. Sci. USA 2002, 99, 371 – 376; d) S.E. Tully, R.
[8] L.-X. Wang, N. Sakairi, H. Kuzuhara, J. Chem. Soc. Perkin Trans.
1 1990, 1677 – 1682.
Mabon, C.I. Gama, S.M. Tsai, X. Liu, L.C. Hsieh-Wilson,
J.
[9] R.R. Schmidt, J. Michel, Angew. Chem. 1980, 92, 763 – 764;
Angew. Chem. Int. Ed. Engl. 1980, 19, 731 – 732.
Am. Chem. Soc. 2004, 126, 7736 – 7737.
[5] a) H.Kunz, H.Waldmann,
Angew. Chem. 1984, 96, 49 – 50;
[10] D.Kumagai, M.Miyazaki, S.I.Nishimura,
Tetrahedron Lett.
Angew. Chem. Int. Ed. Engl. 1984, 23, 71 – 72; b) F.Goto, T.
Ogawa, Tetrahedron Lett. 1992, 33, 5099 – 5102; c) M.Nilsson, J.
Westman, C.-M. Svahn, J. Carbohydr. Chem. 1993, 12, 23 – 37;
d) F.Goto, T.Ogawa, Pure Appl. Chem. 1993, 65, 793 – 801; e) S
Rio, J.-M. Beau, J.-C. Jacquinet, Carbohydr. Res. 1994, 255, 103 –
124; f) J.I.Tamura, J.Nishihara, Bioorg. Med. Chem. Lett. 1999,
9, 1911 – 1914; g) J.I.Tamura, J.Nishihara, J. Org. Chem. 2001,
66, 3074 – 3083.
2001, 42, 1953 – 1956.
[11] B.Thollas, J.C.Jacquinet, Org. Biomol. Chem. 2004, 2, 434 – 442.
[12] a) G.H.Veeneman, S.H.van Leeuwen, J.H.van Boon, Tetra-
hedron Lett. 1990, 31, 1331 – 1334; b) P.Konradsson, UE. .
Udodong, B.Fraser-Reid, Tetrahedron Lett. 1990, 31, 4313 –
4316.
[13] a) M.Kurogochi, T.Matsushita, S.I.Nishimura,
Angew. Chem.
2004, 116, 4163 – 4167; Angew. Chem. Int. Ed. 2004, 43, 4071 –
4075; b) M.Kurogochi, S.I.Nishimura,
6097 – 6101.
[6] a) T.Takamura, S.Tejima, Chem. Pharm. Bull. 1978, 26, 1117 –
1122; b) T.Takamura, T.Chiba, S.Tejima, Chem. Pharm. Bull.
1979, 27, 721 – 725; c) A.Morikawa, H.Kuzuhara, J. Carbohydr.
Chem. 1990, 9, 167 – 179; d) S.I. Nishimura, S. Murayama, K.
Kurita, H.Kuzuhara, Chem. Lett. 1992, 1413 – 1416; e) S.I.
Nishimura, K.Matsuoka, T.Furuike, N.Nishi, S.Tokura, K.
Anal. Chem. 2004, 76,
[14] a) T.Matsushita, H.Hinou, M.Fumoto, M.Kurogochi, N.
Fujitani, H.Shimizu, S.I.Nishimura, J. Org. Chem. 2006, 71,
3051 – 3060; b) K.Naruchi, T.Hamamoto, M.Kurogochi, H.
Hinou, H.Shimizu, T.Matsushita, N.Fujitani, H.Kondo, S.I.
Nishimura, J. Org. Chem. 2006, 71, 9609 – 9621.
Nagami, S.Murayama, K.Kurita,
157 – 163; f) T.Tsuda, T.Furuike, S.I.Nishimura,
Macromolecules 1994, 27,
Bull. Chem.
[15] a) T.Tsuda, S.I.Nishimura, Chem. Commun. 1996, 2779 – 2780;
b) Y.Tachibana, N.Matsubara, F.Nakajima, S.Tsuda, K.
Monde, S.-I. Nishimura, Tetrahedron 2002, 58, 10213 – 10224;
c) Y.Tachibana, G.L.Fletcher, N.Fujitani, S.Tsuda, K.Monde,
S.I. Nishimura, Angew. Chem. 2004, 116, 874 – 880; Angew.
Chem. Int. Ed. 2004, 43, 856 – 862.
Soc. Jpn. 1996, 69, 411 – 416; g) T.Furuike, K.Yamada, T.Ota,
K.Monde, S.I.Nishimura, Tetrahedron 2003, 59, 5105 – 5113.
[7] a) T.Ogawa, T.Nukada, M.Matsui, Carbohydr. Res. 1982, 101,
263 – 270; b) J.Alais, A.Maranduba, A.Veyrieres, Tetrahedron
Lett. 1983, 24, 2383 – 2386.
Angew. Chem. Int. Ed. 2007, 46, 3074 –3079
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3079