Synthesis of the first fully active lipoteichoic acid

Article abstract of DOI:10.1002/anie.200390243

Nature outfoxed! Owing to the sensitivity of lipoteichoic acids (LTAs) to hydrolysis their biological activity could not be ascertained thus far. The chemical synthesis of LTA of Staphylococcus aureus 1, which contains hydrolytically labile D-alanine resi

Full text of DOI:10.1002/anie.200390243

Communications
[9] The hydrogenation of the corresponding Z isomers with the Et-
FerroTANE system leads, as expected, to enantioselectivities of
only 25–80% ee. Orienting investigations with Me-DuPHOS
with the E isomers as substrates gave throughout poorer
enantioselectivities and activities.
pressure)and high substrate/catalyst ratios to very good
enantioselectivities.
Received: September 30, 2002 [Z50265]
[10] In the case of the p-chlorophenyl-substituted product the
determination of the absolute S configuration was achieved by
X-ray structure analysis (see Supporting Information). Since all
other hydrogenation products also gave negative rotations they
should equally be S-configurated, which is in agreement with
literature data.
[11] P. Dierkes, P. W. N. M. van Leeuwen, J. Chem. Soc. Dalton Trans.
1999, 1519 – 1529.
[12] B. Guzel, M. A. Omary, J. P. Fackler, A. Akgerman, Inorg. Chim.
Acta 2001, 325, 45 – 50.
[13] Abreviations used: binapo = 2,2’-Bis(diphenylphosphanyl)-1,1’-
binaphthyl, dipamp = 1,2-ethanediylbis[(2-methoxyphenyl)phe-
nylphosphane], diop = 2,3-O-isopropylidene-2,3-dihydroxy-1,4-
bis(diphenylphosphanyl)butane, Me-DuPHOS = 1,2-bis(2,5-di-
methylphospholanyl)benzene, Et-FerroTANE = 1,1’-bis(2,4-di-
ethylphosphetanyl)ferrocene, nbd = 3,5-norbornadiene, cod =
1,5-cyclooctadiene.
[1] a)E. Juaristi, H. Lopez-Ruiz, Curr. Med. Chem. 1999, 6, 983 –
1004; b) Enantioselective synthesis of b-amino acids (Ed.: E.
Juaristi), Wiley-VCH, Weinheim, 1997; c)G. Cardillo, C. Tom-
asini, Chem. Soc. Rev. 1996, 25, 117 – 128; d)D. C. Cole,
Tetrahedron 1994, 50, 9517 – 9582.
[2] a)D. Heller, H.-J. Drexler, J. You, W. Baumann, K. Drauz, H.-P.
Krimmer, A. Börner, Chem. Eur. J. 2002, 8, 5196 – 5203; b)S.
Lee, Y. J. Zhang, Org. Lett. 2002, 4, 2429 – 2431; c)Y. G. Zhou,
W. Tang, W.-B. Wang, W. Li, X. Zhang, J. Am. Chem. Soc. 2002,
124, 4952 – 4953; d)D. Heller, H.-J. Drexler, A. Spannenberg, B.
Heller, J. You, W. Baumann, Angew. Chem. 2002, 114, 814 – 817;
Angew. Chem. Int. Ed. 2002, 41, 777 – 780; e)D. Heller, J. Holz,
H.-J. Drexler, J. Lang, K. Drauz, H.-P. Krimmer, A. Börner, J.
Org. Chem. 2001, 66, 6816 – 6817; f)M. Yasutake, I. D. Gridnev,
N. Higashi, T. Imamoto, Org. Lett. 2001, 3, 1701 – 1704; g)G.
Zhu, Z. Chen, X. Zhang, J. Org. Chem. 1999, 64, 6907 – 6910;
h)W. D. Lubell, M. Kitamura, R. Noyori, Tetrahedron: Asym-
metry 1991, 2, 543 – 554; i)K. Achiwa, T. Soga, Tetrahedron Lett.
1978, 1119 – 1120; j)B. D. Vineyard, W. S. Knowles, M. J. Sa-
backy, G. L. Bachmann, D. J. Weinkauff, J. Am. Chem. Soc. 1977,
99, 5946 – 5952.
[3] Full characterization and X-ray structural analysis of this
compound, which is reported for the first time, are found in
the Supporting Information.
Synthesis of Lipoteichoic Acids
[4] The hydrogenation of an “E”/Z mixture under normal pressure
at room temperature in MeOH with [Rh(Me-DuPHO-
S)(cod)]BF4[13] leads to the hydrogenation of the Z isomer only,
the “E isomer” remains unchanged. Unlike the starting materi-
als, which are difficult to separate, the hydrogenation product is
readily separated from the unchanged “E isomer” by column
chromatography when the latter is obtained in a pure state.
Synthesis of the First Fully Active Lipoteichoic
Acid**
Andreas Stadelmaier, Siegfried Morath,
Thomas Hartung, and Richard R. Schmidt*
[5] CCDC-194044
(methyl(E)-3-acetylamino-3-phenylacrylate),
CCDC-194047 (methyl 3-diacetylamino-3-phenylacrylate),
Dedicated to Professor Ewald Daltrozzo
on the occasion of his 65th birthday
CCDC-194046 (methyl(S)-3-acetylamino-3-p-chlorophenylpro-
pionate),
CCDC-194045
([Rh{(R,R)-(Et-FerroTA-
NE)}(nbd)]BF4), and CCDC-194043 ([Rh{(S,S)-(Me-Du-
PHOS)}(nbd)]BF4)contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cam-
bridge Crystallographic Data Centre, 12, Union Road, Cam-
bridge CB21EZ, UK; fax: (+ 44)1223-336-033; or deposit@
ccdc.cam.ac.uk).
The inflammatory response to Gram-negative and Gram-
positive bacteria can hardly be distinguished. However, while
most of the responses to Gram-negative bacteria could be
attributed to lipopolysaccharides (LPS)and their lipid anchor
lipid A, as general principles,[1–3] no corresponding priniciple
[6] Compare similar findings on the preparation of analogous but b-
alkyl-substituted substrates: a)D. J. Aberhardt, H.-J. Lin, J. Org.
Chem. 1981, 46, 3749 – 3751; b)P. deMayo, A. C. Weedon, R. W.
Zabel, Can. J. Chem. 1981, 59, 2328 – 2333; c)C. A. Grob, Helv.
Chim. Acta 1950, 33, 1787 – 1796.
[7] a)H.-J. Drexler, W. Baumann, A. Spannenberg, D. Heller, J.
Organomet. Chem. 2001, 621, 89 – 102; b)A. Börner, D. Heller,
Tetrahedron Lett. 2001, 42, 223 – 225; c)C. J. Cobley, I. C.
Lennon, R. McCague, J. A. Ramsden, A. Zanotti-Gerosa,
Tetrahedron Lett. 2001, 42, 7481 – 7483.
[8] a)H. B. Kagan, T.-P. Dang, J. Am. Chem. Soc. 1972, 94, 6429 –
6433; b)W. S. Knowles, M. J. Sabacky, B. D. Vineyard, D. J.
Weinkauff, J. Am. Chem. Soc. 1975, 97, 2567 – 2568; c)M. J.
Burk, J. Am. Chem. Soc. 1991, 113, 8518 – 8519; d)M. J. Burk,
J. E. Feaster, W. A. Nugent, R. L. Harlow, J. Am. Chem. Soc.
1993, 115, 10125 – 10138; e)A. Marinetti, F. Labrue, J.-P. Genet,
Synlett 1999, 12, 1975 – 1977; f)U. Berens, M. J. Burk, A.
Gerlach, W. Hems, Angew. Chem. 2000, 112, 2057 – 2060; Angew.
Chem. Int. Ed. 2000, 39, 1981 – 1984.
[*] Prof. Dr. R. R. Schmidt, A. Stadelmaier
Fachbereich Chemie
Universität Konstanz, Fach M 725
78457 Konstanz(Germany)
Fax: (+49)7531-883-135
E-mail: richard.schmidt@uni-konstanz.de
Dr. S. Morath, Dr. T. Hartung
Fachbereich Biologie
Universität Konstanz
78457 Konstanz(Germany)
[**] This work was supported by the Deutsche Forschungsgemeinschaft
and the Fonds der Chemischen Industrie. We thank S. Deininger
and J. Kley for technical assistance, L. Cobianchi for help in
separating the final products by HI-chromatography, and Prof. Dr.
A. Geyer and A. Friemel for help in the structural assignments of the
final products by NMR experiments.
916
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4208-0916 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 8

Angewandte
Chemie
for Gram-positive bacteria was identified unambiguously
during the last few decades.[4,5] Lipoteichoic acids (LTAs)are
found in most Gram-positive bacteria. Like LPS, they are
amphiphilic, negatively charged glycoplipids.[5] Yet, studies
with commercial LTA preparations exhibiting significant
activities could be traced back to contaminations with
LPS.[6–8] Furthermore, when carefully purified, phenol-ex-
tracted LTA from different sources turned out to be
essentially inactive and cytokine release as a measure of
immunostimulatory activity was not observed.[9–11] Only
recently it was shown that these results are due to the
preparation method employed for LTA:[12] a novel, less harsh
preparation method led to pure, biologically active LTA.[6–8]
By structural analysis a crucial role could be attributed to
preserved d-alanine residues at the polyglycerophosphate
backbone of LTA,[12–14] for instance from Staphylococcus
aureus.
Scheme 1. General structure of lipoteichoic acid (LTA) from Staphylo-
coccus aureus.
To prove the biological activity of d-alanine residues
containing LTA and to allow structure–activity analysis, S.
aureus LTA was synthesized (Scheme 1, A). In this case, the
required combination of glycolipid synthesis with glycero-
phosphate diester formation having varying side chains is
particularly demanding, because the d-alanine residues are
readily cleaved already at pH 8.5. Herein, the first synthesis of
this type of compounds (Scheme 1, with n = 6)is reported. [15]
The versatility of this synthetic concept is further demon-
strated by its application to structural variations. Previous
syntheses of partial LTA structures lacked alanyl residues and
thus did not lead to biologically active materials.[16–18]
of 1a (Scheme 2)which with a 4:1:1 ratio of these residues has
a composition closely related to that of the natural LTA.
Disconnection of 1a between the glycerophosphate backbone
and gentiobiosyl diacylgycerol (see arrows in 1a)leading to
intermediates 2 and 3 and N-benzyloxycarbonyl(Z)-protected
d-alanine should provide a viable approach to the synthesis of
the target molecule. Thus, 2 should be accessible from
glycerol derivatives 4–6 as building blocks, and 3 from the
readily available gentiobiosyl diacylglycerol derivative 7. The
choice of permanent and temporary orthogonal protecting
groups was clearly decisive for success: for chain extension of
the glycerophosphate backbone and the final ligation be-
tween intermediates 2 and 3 the tert-butyldiphenylsilyl
(TBDPS)group was selected, and for the execution of the
The approximate ratio of the glycerophosphate 2-O-
substituents in A (Scheme 1)of 70% d-alanyl, 15% a-d-
GlcNAc, and 15% hydrogen[12] led us to design the synthesis
Scheme 2. Structure of target molecule 1a and its retrosynthesis. Bn=benzyl.
Angew. Chem. Int. Ed. 2003, 42, No. 8
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 1433-7851/03/4208-0917 $ 20.00+.50/0
917

Communications
synthesis of 3 the isopropylidene group was chosen. The
selective attachment of the alanyl residues was based on the
mild oxidative removal of the methoxybenzyl (MPM)group,
thus leading to the fully protected target molecule having only
benzyl and Z-protecting groups which should finally be
removable by hydrogenolysis. The variability of this concept
in terms of different combinations of building blocks 4–6,
attachment of different amino acid residues, and variation of
[19]
the diacylglycerol moiety led to several derivatives of 1a,
for example 1b with l-alanyl residues.
The required building block 4 was obtained from com-
mercially available 1,2-O-isopropylidene-sn-glycerol.[20] Then
O-allylation de-O-isopropylidenation[21] and regioselective
introduction of the TBDPS group gave intermediate 8
(Scheme 3)which was used for the attachment of the MPM
group. De-O-allylation of this MPM-protected intermediate,
and then reaction with benzyloxybis(diisopropylamino)phos-
phane[22] in the presence of tetrazole afforded building block 5
(Scheme 2).
Glycosylation of intermediate 8 with known azidoglucosyl
donor 9[23] led under trimethylsilyl trifluoromethanesulfonate
(TMSOTf)catalysis to a-connected glucoside 10 in excellent
75% yield (Scheme 3). Replacement of the O-acetyl groups
by benzyl groups (!11)and then reduction of the azido
group and N-acetylation under standard conditions[24] afford-
ed 12. De-O-allylation and then phosphitylation as described
above furnished building block 6.
Scheme 4. Synthesis of intermediate 2.
of tetrazole and then addition of tert-butyl hydroperoxide as
oxidizing agent afforded a phosphorus triester intermediate,
which on successive treatment with tetrabutylammonium
fluoride (TBAF)and reaction with 5 led to glycerophosphate
intermediate 13. Its reaction with TBAF, then reaction with 6
instead of 5, and again treatment with TBAF furnished the
desired compound 2 in very good overall yield.
For the synthesis of 3, gentiobiose was transformed under
standard conditions (four steps)into glycoside 14 (Scheme 5).
Removal of the O-benzoyl groups under ZemplØn conditions,
then regioselective silylation at the 6b oxygen atom and per-
O-benzylation afforded building block 7. Acid-catalyzed de-
O-isopropylidenation, then acylation with myristoyl chloride
in the presence of triethylamine followed by desilylation gave
intermediate 15 which on phosphitylation as described above
furnished the desired compound 3; on hydrogenolytic de-O-
benzylation gentiobiosyl diacylglycerol 16 was obtained. The
latter compound was employed to study the biological effect
of the lipid anchor having only the gentiobiose core.
With building blocks 4–6 in hand, intermediate 2 could be
constructed (Scheme 4). Reaction of 4 with 5 in the presence
Scheme 5. Synthesis of intermediate 3 and of gentiobiosyl diacylglycer-
ol 16. a) 1. BzCl, pyridine, 408C, 24 h, 98%; 2. N2H4·HOAc, DMF,
508C, 3.5 h, 78%; CCl3CN, DBU, CH2Cl2, 2 h, 97%; 3. BF3·OEt2
(0.2 equiv), 3-Š MS, À308C, CH2Cl2, 1.2-O-isopropylidene-sn-glycerol,
80%; b) 1. NaOMe, MeOH, 24 h; TBDPS-Cl, pyridine, À158C, two
days, 91%; 2. BnBr, NaH, DMF, 66%; c) 1. HOAc (75%), 808C, 1.5 h,
78%; 2. myristoyl chloride, NEt3, THF, 508C, 4 h, 81%; 3. TBAF,
HOAc, THF, 408C, three days, 76%; d) Pd/C, H2, CH2Cl2/MeOH (2:5),
24 h, 71%; e) benzyloxybis(diisopropylamino)phosphane, CH2Cl2,
79%. Bz=benzoyl.
Scheme 3. Synthesis of building block 6. a) TMSOTf, CH2Cl2, RT, 75%;
b) 1. NaOMe, MeOH, 24 h; 2. BnBr, NaH, DMF, 24 h (74%); c) 1. 1,3-
propanedithiol, pyridine, H2O, RT, 24 h; 2. pyridine/Ac2O, 2 h, 85%;
d) 1. [(PPh3)3RhCl], DBU, EtOH, 908C; 2. 1m HCl/actone (1:9), 77%;
3. benzyloxybis(diisopropylamino)phosphane, CH2Cl2, 85%.
DBU=1,8-diazabicyclo[5.4.0]undec-7-ene, All=allyl.
918
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4208-0918 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 8

Angewandte
Chemie
The ligation of intermediate 2 and 3 could be performed
with tetrazole as catalyst; oxidation with tert-butyl hydro-
peroxide furnished 17, which contains the backbone of the
target molecule (Scheme 6). Treatment with ceric(iv)ammo-
In conclusion, with the help of synthetic material the
active principle of LTA stimulation of the immune system
could be ascertained for the first time. The synthesis design
presented considers the variable substitution pattern and the
hydrolytic lability of the alanyl residues of the glycer-
ophosphate backbone. It is also amenable to the
combinatorial synthesis of a great variety of LTAs.
Received: September 20, 2002 [Z50223]
[1] J. B. Zivot, W. D. Hoffmann, New Horiz. 1995, 3, 267 –
275.
[2] E. T. Rietschel, H. Brade, O. Holst, L. Brade, S. Müller-
Loennies, U. Mamat, U. Zähringer, F. Beckmann, U.
Seydel, K. Brandenburg, A. Ulmer, T. Mattern, H. Heine,
J. Schletter, S. Hauschildt, H. Loppnow, U. Schönbeck,
H.-D. Flad, U. F. Schade, F. DiPadova, S. Kusumoto,
R. R. Schumann, Curr. Top. Microbiol. Immunol. 1996,
216, 39 – 81.
[3] U. Seydel, M. Oikawa, K. Fukase, S. Kusumoto, K.
Brandenburg, Eur. J. Biochem. 2000, 267, 3032 – 3039, and
references therein.
Scheme 6. Synthesis of target molecule 1a and its diastereoisomer 1b. a) Tet-
razole, CH2Cl2; tBuO2H, 75%; b) CAN, MeCN/toluene/H2O (60:3:4), 45–
60 min, 67%; c) 1. PyBOP (20 equiv), N-methylimidazole (40 equiv), Z-protect-
ed d/l-Ala (HNEt3+ salt) (20 equiv), CH2Cl2, 3 h; d) Pd(OH)2/C, H2, CH2Cl2/
MeOH/H2O=(5:5:1), 24 h; HI chromatography.
[4] S. M. Opal, J. Cohen, Crit. Care Med. 1999, 27, 1608 –
1616.
[5] W. Fischer, Med. Microbiol. Immunol. 1994, 183, 61 – 76.
[6] J. J. Gao, Q. Xue, E. G. Zuvanich, K. R. Haghi, D. C.
Morrison, Infect. Immun. 2001, 69, 751 – 757.
nium nitrate (CAN)liberated four of the glycerol hydroxy
groups, to which the Z-protected d-alanyl residues were
immediately attached with (benzotriazol-1-yloxy)tripyrroli-
dinophosphonium hexafluorophosphate (PyBOP)/N-methyl-
imidazole as condensing agent. Hydrogenolysis of this
intermediate with Pearlman's catalyst[25] in a mixture of
CH2Cl2/MeOH/H2O (5:5:1)gave the desired final product 1a
after hydrophobic interaction (HI)chromatography on oc-
tylsepharose in 47% yield. The structural assignment was
based on MS and NMR data[26] and comparison with naturally
occuring material. Compound 17 was similarly transformed
with Z-protected l-alanine into 1b.
[7] N. Kawamura, N. Imanishi, H. Koike, H. Nakahara, L. Phillips, S.
Morooka, Biochem. Biophys. Res. Commun. 1995, 217, 1208 –
1215.
[8] A. Sugiyama, R. Arakaki, T. Ohnishi, N. Arakaki, Y. Daikuhara,
H. Takada, Infect. Immun. 1996, 64, 1426 – 1431.
[9] R. Keller, W. Fischer, R. Keist, S. Bassetti, Infect. Immun. 1992,
60, 3664 – 3672.
[10] T. Kusunoki, E. Hailman, T. S. Juan, H. S. Lichenstein, S. D.
Wright, J. Exp. Med. 1995, 182, 1673 – 1682.
[11] Y. Suda, H. Tochio, K. Kawano, H. Takada, T. Yoshida, S.
Kotani, S. Kusumoto, FEMS Immunol. Med. Microbiol. 1995, 12,
97 – 112.
The evaluation of the biological activity of compounds 1a,
1b, and 16 revealed that 1a with d-alanyl residues (Figure 1)
has practically the same activity in terms of initiation of
cytokine release by human blood leukocytes as the natural
LTA. Whereas the lipid anchor with the gentiobiose core
alone (compound 16)shows practically no activity, l-alanine-
containing 1b was about 10- to 100-fold less active than
[12] S. Morath, A. Geyer, T. Hartung, J. Exp. Med. 2001, 193, 393 –
397.
[13] S. Morath, A. Stadelmaier, A. Geyer, R. R. Schmidt, T. Hartung,
J. Exp. Med. 2002, 195, 1635 – 1640.
[14] S. Deininger, A. Stadelmaier, S. von Aulock, S. Morath, R. R.
Schmidt, T. Hartung, J. Immunol. submitted.
[15] Some biological data of these investigations have already been
reported: see ref. [13].
[16] J. J. Oltvoort, C. A. A. van Boeckel, J. H. De Koning, J. H.
van Boom, Recl. Trav. Chim. Pays-Bas 1982, 101, 87 – 91; J. J.
Oltvoort, M. Kloosterman, C. A. A. van Boeckel, J. H. van -
Boom, Carbohydr. Res. 1984, 130, 147 – 163.
[13,14]
1a.
[17] K. Fukase, T. Matsumoto, N. Ito, T. Yoshimura, S. Kotani, S.
Kusumoto, Bull. Chem. Soc. Jpn. 1992, 65, 2643 – 2654; K.
Fukase, T. Yoshimura, S. Kotani, S. Kusumoto, Bull. Chem. Soc.
Jpn. 1994, 67, 473 – 482; S. Kusumoto, K. Fukase, Y. Suda, M.
Oikawa, J. Synth. Org. Chem. Jpn. 1996, 54, 976 – 987.
ˇ
ˇ
´
´
ˇ ´
[18] I. Jeric, L. Simicic, M. Stipetic, S. Horvat, Glycoconjugate J. 2000,
17, 273 – 282.
[19] A. Stadelmaier, Dissertation, University of Konstanz, submitted.
[20] B. Wickberg, Acta Chem. Scand. 1958, 12, 1187 – 1201.
[21] P. A. Gent, R. Gigg, J. Chem. Soc. Perkin Trans. 1 1975, 364 – 370.
[22] W. Bannwarth, A. Trzeciak, Helv. Chim. Acta 1987, 70, 175 – 186.
[23] G. Grundler, R. R. Schmidt, Liebigs Ann. Chem. 1984, 1826 –
1847.
Figure 1. Concentration dependence of TNFa release of diluted human
whole blood (20%) in response to 1a, 1b, and 16. TNFa was meas-
ured by ELISA. Data are means (Æ SEM) of twelve donors. SEM=
standard error of the mean.
Angew. Chem. Int. Ed. 2003, 42, No. 8
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 1433-7851/03/4208-0919 $ 20.00+.50/0
919

Communications
[24] H. Bayley, D. N. Standring, J. R. Knowles, Tetrahedron Lett.
calcd: C 66.40, H 6.86, N 0.35; found: C 66.40, H 7.09, N 0.36. 1a:
[a]2D0 = + 6.9 (c = 0.1, H2O); 1H NMR (600 MHz, D2O): d = 0.82–
0.94 (m, 6H; Me), 1.16–1.43 (m, 40H; CH2(g-w)), 1.55–1.69 (m,
16H; Ala-bH, CH2(b)), 2.11 (s, 3H; NHAc), 2.29–2.47 (m, 4H;
CH2(a)), 3.27–3.73, 3.75–4.34, 4.36–4.59 (m, 54H), 5.10 (bs, 1H;
1c-H), 5.31–5.43 ppm (m, 5H; 5, 8, 11, 14-H, 2’-H); 13C NMR
(150.8 MHz, D2O): d = 14.2 (Me), 15.9 (Ala-Me), 22.6 (NHAc),
25.4 (CH2(b)), 30.5 (CH2(g-w)), 34.5 (CH2(a)), 49.2 (CHNH3+),
54.0 (2c-C), 60.9 (6c-C), 62.4 (18-C), 64.0 (CH2-Glyc.-C, 3’-C),
65.5 (1, 3-C), 66.8 (16-C), 68.2 (1’-C), 70.3 (4c-C), 71.1 (17-C),
71.4 (3c-C), 72.4 (5c-C), 73.4 (2a, 2b-C), 74.4 (5, 8, 11, 14-C), 75.9
(3a, 3b-C), 76.2 (2-C), 97.2 ppm (1c-C); MALDI-TOF MS
(negative mode, matrix: trihydroxyacetophenone/diammonium
citrate 2:1): [MÀH]À, calcd: m/z: 2247.9, found: m/z: 2248.5;
1978, 3633 – 3634.
[25] L. F. Fieser, M. Fieser, Reagents for Organic Synthesis, Vol. 1,
Wiley, New York, 1967, p. 782.
[26] NMR experiments were performed at 600 MHz and 300 K. The
NMR spectra of the final compounds were related to 3-
(trimethylsilyl)3,3,2,2-tetradeuteropropionic acid Na salt
([D4]TSPA). 13C Assignments were based on heteronuclear
multiple-quantum correlation (HMQC). The anomeric proton
of N-acetylglucosamine in compound 10 showed a doublet with a
vicinal coupling constant of 3.6 Hz as expected for an a-
glycosidic bond. The protons of the anomeric center gentiobiose
intermediate 14 resolved also to doublets with coupling con-
stants of 7.9 Hz and 7.8 Hz as expected for b-glycoside bonds. 17:
TLC (toluene/acetone 1:1): Rf = 0.4; [a]2D0 = + 12.1 (c = 1,
CHCl3); 1H NMR (600 MHz, CDCl3, TMS): d = 0.81–0.92 (t,
6H; me), 1.11–1.35 (m, 40H; CH2(g-w)), 1.44–1.61 (m, 4H;
CH2(b)), 1.84–1.98 (m, 3H; NHAc), 2.09–2.25 (m, 4H; CH2(a)),
3.32 (m, 1H; 5a/b-H), 3.33 (m, 1H; 2a-H), 3.35 (m, 1H; 1’-H),
3.37 (m, 1H; 5a/b-H), 3.38 (m, 1H; 2b-H), 3.43, 3.44 (m, 2H; 4a,
4b-H), 3.49 (m, 2H; 18-H), 3.57 (m, 2H; 3a, 3b-H), 3.58–3.60 (m,
3H; 6c, 6a-H), 3.63(m, 4H; 5, 8, 11, 14-H), 3.64 (m, 1H; 4c-H),
3.67 (m, 12H; OMe), 3.68 (m, 1H; 3c-H), 3.69 (m, 1H; 2-H), 3.72
(m, 1H; 17-H), 3.75 (m, 1H; 5c-H), 3.83 (m, 1H; 1’-H), 3.96, 4.04
(m, 18H; 3, 4, 6, 7, 9, 10, 12, 13, 15-H), 4.06, 4.15 (m, 2H; 3’-H),
4.08 (m, 1H; 6a-H, 4.07, 4.17(m, 4H; 16, 6b-H), 4.20 (m, 1H; 1a-
H), 4.35 (m, 1H; 2c-H), 4.41 (m, 1H; 1b-H), 4.74 (m, 1H; 1c-H),
4.29–4.91 (m, 30H; CH2Ph), 4.93 (m, 12H; POCH2Ph), 5.06 (m,
1H; 2’-H), 6.69–6.80 (m, 8H; PhMPM), 7.03–7.34 ppm (m, 93H;
Ph); 13C NMR (150.8 MHz, CDCl3, TMS): d = 14.11 (2C; me),
23.1 (1C; NHAc), 24.84 (2C; CH2(b)), 22.68/29.14–31.91 (20C;
CH2(g-w)), 34.04, 34.22 (2C; CH2(a)), 52.8 (1C; 2c-C), 55.17
(4C; OMe), 63.2 (1C; 3’-C), 65.5–67.2 (12C; CH2-Glyc., 6b-C),
68.1 (1C; 1’-C), 68.5 (1C; 6c-C), 68.6 (1C; 6a-C), 69.0 (1C; 18-C)
69.5 (6C; POCH2Ph), 69.8 (1C; 2’-C), 72 (1C; 5c-C), 73.5 (1C;
5a/b-C), 75 (1C; 4a/b-C), 75.4 (4C; 5, 8, 11, 14-C), 72–77.0 (15C;
CH2Ph), 76.8 (1C; 17-C), 77.5 (1C; 4c-C), 78.2 (1C; 2-C), 81.2
(1C; 3c-C), 81.8 (2C; 2a, 2b-C), 84.6 (2C; 3a, 3b-C), 100.05 (1C;
1c-C), 103.1 (1C; 1a-C), 103.6 (1C; 1a-C), 103.6 ppm (1C; 1b-
C); assignment according to 1a in Scheme 2; MALDI-TOF MS
(positive mode, matrix: p-nitroaniline + NaI, THF): [MþNa]+,
calcd: m/z 4002.3; found: m/z 4000.0; C220H271NO54P6 (3979.3):
[(M-Ala]À, calcd: m/z: 2176.9, found: m/z: 2177.6. 1b: [a]2D0
=
+ 5.1 (c = 0.2, H2O); NMR data are almost identical with those
of diastereomer 1a; MALDI-TOF MS (negative mode, matrix:
trihydroxyacetophenone/diammonium citrate 2:1): [MÀH]À,
calcd: m/z: 2247.9, found: m/z: 2245.5; [(MÀAla)]À, calcd:
m/z: 2176.9, found; m/z: 2174.8; (positive mode, matrix: DHB):
[MþH]+, calcd: m/z: 2249.9, found: m/z: 2249.9.
[27] Whole blood cytokine response: Differential blood cell counts
were routinely performed with a Pentra 60 analyzer (ABX
Diagnostics, Montpellier, France)to exclude acute infections of
blood donors. Heparinized blood (200 mL)freshly taken from
healthy volunteers was diluted fivefold with isotonic sodium
chloride solution (0.9%, Boehringer Ingelheim, Germany). LTA
was sonified before use to dissolve micelles. After addition of the
LTA stimuli in a volume of 10 mL, incubations were carried out
in polypropylene reaction tubes (Eppendorf, Hamburg, Germa-
ny)in the presence of 5% CO at 378C overnight. Cell-free
2
supernatants were obtained by centrifugation at 400 ꢀ g for 2 min
and stored at À708C for cytokine determination. Cytokines were
measured by sandwich ELISA (enzyme-linked immunosorbent
assay)based on an antibody pair against human TNF
a
(Endogen, Eching, Germany). Recombinant human TNFa
serving as the standard was a gift from Dr. S. Poole, National
Institute for Biological Standards and Controls, London (UK).
Binding of biotinylated antibody was quantified by using
streptavidin peroxidase (Biosource, Camarillo, CA. USA)and
the substrate TMB (3,3’,5,5’-tetramethylbenzidine, Sigma,
Deisenhofen, Germany).
920
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4208-0920 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 8