Total synthesis of the glycopeptide recognition domain of the P-selectin glycoprotein ligand 1

Article abstract of DOI:10.1002/anie.200705762

(Chemical Presented) Block glycosylation of a T antigen-threonine conjugate with a sialyl Lewis^x trichloroacetimidate was used to synthesize a sialyl Lewis^x-T antigen-threonine building block. Selective protecting-group manipulations furnished a compound with exclusively acetyl-protected functions in the carbohydrate portion; this intermediate is sufficiently acid-stable that it is applicable to the solid-phase synthesis of binding site A of the P-selectin ligand PSGL-1.

Full text of DOI:10.1002/anie.200705762

Angewandte
Chemie
DOI: 10.1002/anie.200705762
Total Synthesis
Total Synthesis of the Glycopeptide Recognition Domain of the
P-Selectin Glycoprotein Ligand 1**
Katharina Baumann, Danuta Kowalczyk, and Horst Kunz*
During the recruitment of leukocytes into inflamed tissues in
acute and chronical inflammatory processes, the carbohy-
drate-recognizing receptors P- and E-selectin exposed on the
activated endothelium and L-selectin on the leukocytes play
most important roles.[1] A similar situation holds true for the
metastasis of tumor cells.[2] The intrusion of malign T cells
into the skin could be more readily influenced through ligands
of P-selectin than ligands of E-selectin. Therefore, in this case
the inhibition of P-selectin provided the more promising
antitumor strategy.[3] In contrast to the natural ligands of E-
selectin,[4] it is known for the P-selectin glycoprotein ligand 1
(PSGL-1) that in addition to the essential tetrasaccharide
sialyl Lewisx the N-terminal peptide sequence significantly
contributes to the recognition epitope.[5] Within the PSGL-1
the binding site for binding to P-selectin has been identified
by biochemical and molecular biological methods (Figure 1).
by enzymatic glycosylation reactions. Cummings et al.[7]
conducted the chemical solid-phase synthesis of the glyco-
peptide Gly41–Leu63 carrying an aGalNAc residue, and they
subsequently performed all glycosylation reactions and the
final sulfation using the corresponding enzymes. They report
a high affinity of the glycopeptide ligand, which was obtained
in micrograms quantities, to P-selectin (Kd = 350 nm). This
affinity nearly equalled that measured for PSGL-1isolated
from neutrophiles (300 nm). These results give evidence that
the recognition site of PSGL-1constitutes a particularly
interesting structure for detailed model studies of cell
adhesion phenomena. This holds even more so, as it was
recently found that the bacterium Anaplasmosis phagocyto-
phylum, which causes human granulocytoplasmosis (Ehrli-
chiosis), expresses a receptor analogous to P-selectin which
[8]
binds the PSGL-1ligand in the non-sulfated form.
In
addition, it was shown that the separate presentation of
x
partial structures of PSGL-1as sialyl Lewis and O-sulfated
tyrosine in conjugates of polyacrylamide can result in a high
affinity to L-selectin.[9] Considering these significant functions
of the natural P-selectin ligand in the regulation of cell
adhesion and in their aberrant forms causing severe diseases,
we deemed that a chemical synthesis of the recognition
domain of PSGL-1on a preparative scale would be desirable.
Such a synthesis would provide sufficient amounts of exactly
specified compounds for model investigations of cell adhesion
processes and for enzymatic modifications, for example the
O-sulfation of tyrosine, as well as for structural investigations.
Furthermore, the chemical synthesis would pave the way for
the preparation of structural mimics that are more resistant
towards biological degradation, as was recently shown for
glycopeptide ligands of E-selectin.[10]
Figure 1. Binding site of P-selectin glycoprotein ligand 1 (PSGL-1) for
binding to P-selectin.
It is located in the N-terminal portion, which, after cleavage
of a signal peptide, begins with Gln42. At least one of the
tyrosine residues Tyr46, Tyr48, and Tyr51 should be O-sulfated.
The threonine in position 57 carries the O-glycan side chain,
which consists of a combination of a sialyl Lewisx ligand and
the core 2 structure of O-glycoproteins (mucins).
The synthesis of the protected sialyl Lewisx–core2–threo-
nine building block I certainly is the most challenging
problem in the chemical construction of the recognition site
of PSGL-1. According to the retrosynthetic analysis
(Scheme 1) I should be accessible from the sialyl Lewisx
donor II and the partially deprotected T antigen–threonine
conjugate III, which is known from syntheses of tumor-
associated glycopeptide antigens.[11]
Chemoenzymatic syntheses of partial sequences[6] and of
the entire binding site of PSGL-1[7] have been described.
Wong et al.[6] synthesized the glycopeptide Tyr51–Glu58 which
was O-glycosylated at Thr57 with the disaccharide GlcNAcb-
(1–6)-aGalNAc. After chemical sulfation of the tyrosine, the
sialyl Lewisx structure was assembled at the GlcNAc residue
The N-protecting group of the glucosamine is decisive for
the b-selective block glycosylation of acceptor III with donor
II. Based on previous experience,[12] the 2,2,2-trichloroethoxy-
carbonyl (Troc) group was applied; it should prevent the
formation of a less reactive oxazoline. Sufficient reactivity of
the donor should be guaranteed by the trichloroacetimidate
group according to Schmidt and Michel.[13a] The regioselective
glycosylation at the 6-OH group can be expected in agree-
ment with numerous previous examples.[11,14] The axial 4-OH
group is of low reactivity, in particular, in molecules carrying a
[*] Dr. K. Baumann, D. Kowalczyk, Prof. Dr. H. Kunz
Institut fuer Organische Chemie
Johannes Gutenberg-Universitaet Mainz
Duesbergweg 10–14, 55128 Mainz (Germany)
Fax: (+49)6131-392-4786
E-mail: hokunz@mail.uni-mainz.de
[**] This work was supported by the Deutsche Forschungsgemeinschaft.
K.B. is grateful for a PhD scholarship provided by the Fonds der
Chemischen Industrie.
Angew. Chem. Int. Ed. 2008, 47, 3445 –3449
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3445

Communications
Scheme 2. Synthesis of the sialyl Lewisx glycosyl donor. DMAP=4-
dimethylaminopyridine, TMSOTf=trimethylsilyl trifluoromethansulfo-
nate.
Scheme 1. Retrosynthesis of the sialyl Lewisx—Tantigen–threonine
building block I. Bn=benzyl, Fmoc=9-fluorenylmethoxycarbonyl,
Troc=trichlorethoxycarbonyl, TBDPS=tert-butyldiphenylsilyl.
large substituent at O-3. The protecting group at the anomeric
oxygen of the glucosamine building block IV1 is also of
importance. This group must be stable under acidic (during
glycosylation reactions) and basic conditions (during removal
of O-acetyl groups). However, finally it must be selectively
removable and, thus, make it possible to introduce the
trichloroacetimidate, giving II. The tert-butyldiphenylsilyl
group[15] was supposed to meet these requirements. The
other monosaccharide building blocks IV2–IV4 have already
been used successfully in syntheses of sialyl Lewisx deriva-
tives.[10]
tetrabutylammonium fluoride (TBAF) in tetrahydrofuran/
acetic acid (100:1). Subsequent reaction with trichloroaceto-
nitrile[13] in the presence of 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU) resulted in formation of sialyl Lewisx trichloroaceti-
midate II (Scheme 3). Both reactions proceeded slowly with
the sterically demanding substrate.
For the construction of the sialyl Lewisx donor II
(Scheme 2), peracetylated N-trichloroethoxycarbonyl glucos-
amine 1 was deacetylated at O-1by treatment with hydrazine
acetate.[12,16] The product was subsequently silylated with tert-
butyldiphenylsilyl chloride (TBDPS-Cl)[15] to give 2. After
removal of the O-acetyl groups by mild transesterification
with catalytic NaOMe in methanol, the 4,6-O-benzylidene
acetal was introduced. Product 3 served as the glycosyl
acceptor in the reaction with thiofucoside[17] 4 under modified
in situ anomerization conditions.[18] Regioselective opening of
acetal 5 was performed with triethylsilane/trifluoromethane-
sulfonic acid[19] to give 6, which was subsequently glycosylated
with galactosyl trichloroacetimidate[13] 7 to furnish the Lewisx
trisaccharides 8.
Scheme 3. Block glycosylation of a T antigen–threonine conjugate with
a sialyl Lewisx trichloroacetimidate.
The O-acetyl groups were removed from 8 by mild
transesterification in methanol (see above). The sialic acid
moiety was regio- and stereoselectively introduced to the
resulting Lewisx derivative 9 according to a previously
described method[10,11,20] using xanthate 10.[21] Acetylation of
the 2- and 4-OH groups of the galactose portion furnished
sialyl Lewisx tetrasaccharide 11. Remarkably, the TBDPS
group remained unaffected during all these conversions
(Scheme 2).
The block glycosylation of the partially deprotected
bGal(1–3)aGalNAc–threonine derivative III,[11] carrying the
Fmoc/tBu ester protecting-group combination, with SLex
trichloroacetimidate II constitutes the key step of the overall
synthesis. It led to products I and 12, which have identical
molecular weight, in a total yield of more than 80%. The
isomers could be separated by flash chromatography. Two-
dimensional NMR spectroscopy (HSQC and HMQC spectra)
revealed, in particular through the position of the 13C signal
for C-6 of the GalNAc moiety (d = 69.2 ppm for I, d =
59.2 ppm for 12), that the major product 12 had been
In order to transform tetrasaccharide 11 into a glycosyl
donor, the TBDPS group was removed by treatment with
3446 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3445 –3449

Angewandte
Chemie
formed by an entirely unexpected glycosylation at the axial 4-
OH group of T antigen–threonine derivative III. This regio-
selectivity is in contradiction to every previous observa-
tion[11,14,20] and shows that unexpected effects can arise in the
course of slowly proceeding glycosylation reactions between
large building blocks. Fortunately, the desired compound I
was isolated in a sufficient amount (0.66 g) so that the
synthesis of the glycopeptide of the recognition domain
(Figure 1) could be continued.
activation of the demanding Fmoc-glycosyl–threonine 15
(Scheme 5). Coupling of 15 required a reaction time of 8 h.
The amino acids Glu, Pro, and Leu following the glycosyl
threonine were attached in a double coupling reaction (one
repetition). After completion of the glycopeptide sequence
Starting from I, an Fmoc-protected glycosyl threonine
building block is not directly accessible since its a-fucoside
structure is sensitive even to dilute trifluoracetic acid in
dichloromethane. It is indispensable to stabilize this structure
towards acids by an exchange of the O-benzyl for the O-acetyl
protecting groups.[10,22] Fist, however, the removal of the N-
Troc group had to be carried out by reductive elimination with
zinc,[12] otherwise, the Troc group would lose chloro substitu-
ents during the hydrogenation reaction. After N-acetylation,
the benzyl ethers of product 13 were hydrogenolytically
cleaved. Monitoring this conversion indicated that a simulta-
neous loss of the Fmoc protecting group occurred. Therefore,
selective re-introduction of the Fmoc group using Fmoc-O-
hydroxysuccinimide (Fmoc-OSu) was necessary prior to the
O-acetylation to give 14 (Scheme 4).[23]
Scheme 5. Solid-phase synthesis of the O-glycopeptide of the P-selec-
tin binding site of PSGL-1. [a] Solid-phase synthesis cycle: 1) Fmoc
removal: piperidine/NMP (1:4); 2) amino acid coupling: Fmoc-AA-OH,
HBTU, HOBt, DIPEA, DMF; in the case of 15: HATU, HOAt, NMM,
NMP; 3) Capping: Ac2O, DIPEA, HOBt, NMP.
Scheme 4. Protecting-group manipulations giving an acid-stable Fmoc-
glycosyl–threonine building block for the solid-phase synthesis.
and final N-acetylation, the linker was cleaved with acetic
acid/trifluoroethanol in dichloromethane[25] without affecting
the tert-butyl esters and ethers. These groups were subse-
quently removed by treatment with trifluoroacetic acid to
yield glycopeptide 17. Preparative HPLC was employed to
separate the C-terminal dipeptide and the glycosyltripeptide
as the by-products from 17, which was isolated as the main
fraction (23%) besides compounds identical in molecular
weight and showing very similar retention times (17%). Since
re-injection of these compounds gave an HPLC profile very
similar to that of crude 17, they are probably conformers or
different salt forms of 17. The main fraction was treated with
catalytic NaOMe in methanol (pH > 9), but this did not result
in complete removal of the O-acetyl groups. During the
subsequent hydrolysis of the sialic acid methyl ester with
NaOH in water at pH 10.5[20] (the pH has to be maintained
carefully, otherwise b elimination of the entire glycan may
occur), the remaining O-acetyl groups were removed. After
neutralization with acetic acid, the pure sialyl Lewisx–
T antigen–glycopeptide 18[28] of the binding site of P-selectin
Compound 14 is sufficiently stable towards acids, and its
tert-butyl ester can be cleaved selectively with trifluoroacetic
acid (TFA). The Fmoc(sialyl Lewisx–T antigen)–threonine
building block 15, suitable for the solid-phase synthesis of the
glycopeptide was obtained in a pure form. Compound 15 was
used for the solid-phase synthesis of the sequence Tyr48–Pro59
of PSGL-1which was performed according to the Fmoc
strategy starting from polystyrene resin 16,[24] which was
loaded with proline attached by means of the Barlos linker.[25]
For Fmoc-amino acids, O-(benzotriazol-1-yl)-N,N,N’,N’-
tetramethyluronium hexafluorophosphate (HBTU), 1-
hydroxybenzotriazole (HOBt),[26] and diisopropylethylamine
(DIPEA) in DMF served as coupling reagents, whereas the
more reactive mixture O-(7-azabenotriazol-1-yl)-N,N,N’,N’-
tetramethyluronium hexafluorophosphate (HATU), 7-aza-1-
hydroxybenzotriazol (HOAt),[27] and N-methylmorpholine
(NMM) in N-methylpyrrolidinone (NMP) was applied for
Angew. Chem. Int. Ed. 2008, 47, 3445 –3449
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3447

Communications
[8] T. Yago, A. Leppänen, J. A. Carlyon, M. Akkoyunlu, S.
Karmakar, E. Fikrig, R. D. Cummings, R. P. McEver, J. Biol.
Chem. 2003, 278, 37987.
[9] S. Enders, G. Bernhard, A. Zakrzewicz, R. Tauber, Biochim.
Biophys. Acta Gen. Subj. 2007, 1770, 1441.
[10] a) C. Filser, D. Kowalczyk, C. Jones, M. K. Wild, U. Ipe, D.
Vestweber, H. Kunz, Angew. Chem. 2007, 119, 2155; Angew.
Chem. Int. Ed. 2007, 46, 2108; b) U. Sprengard, G. Kretzschmar,
E. Bartnik, C. Hüls, H. Kunz, Angew. Chem. 1995, 107, 1104;
Angew. Chem. Int. Ed. Engl. 1995, 34, 990.
[11] a) C. Brocke, H. Kunz, Synthesis 2004, 525; b) S. Dziadek, D.
Kowalczyk, H. Kunz, Angew. Chem. 2005, 117, 7798; Angew.
Chem. Int. Ed. 2005, 44, 7624.
ligand PSGL-1was isolated as 31mg of a colorless lyophili-
sate.
In the total synthesis of PSGL-1binding epitope 18,
problems accumulate that complicate the chemical glycopep-
tide synthesis. In order to stereoselectively install the a-
fucoside structure, a benzyl ether protected fucosyl donor
must be used. However, the resulting fucoside bond is acid-
labile and not compatible with the conditions needed for the
cleavage of tert-butyl groups during the solid-phase synthesis.
In turn, the necessary exchange of the O-benzyl for O-acetyl
protecting groups does not allow protection of the sialic acid,
whose chemistry already is distinctly difficult, as a benzyl
ester.[29] Hence, the sialic acid methyl ester must be hydro-
lyzed after the long synthesis in basic media without effecting
a b elimination of the entire glycan.[30]
[12] a) M. Schultz, H. Kunz, Tetrahedron Lett. 1992, 33, 5319; b) M.
Schultz, H. Kunz, Tetrahedron: Asymmetry 1993, 4, 1205; c) P.
Boullanger, M. Jouineau, B. Bouammali, D. Lafont, G. Descotes,
Carbohydr. Res. 1990, 202, 151.
[13] a) R. R. Schmidt, J. Michel, Angew. Chem. 1980, 92, 763; Angew.
Chem. Int. Ed. Engl. 1980, 19, 731; b) W. Stahl, U. Sprengard, G.
Kretzschmar, D. W. Schmidt, H. Kunz, J. Prakt. Chem. 1995, 337,
441.
[14] a) A. Kuhn, H. Kunz, Angew. Chem. 2007, 119, 458; Angew.
Chem. Int. Ed. 2007, 46, 454; b) B. Liebe, H. Kunz, Tetrahedron
Lett. 1994, 35, 8777.
[15] S. Hanessian, P. Lavallee, Can. J. Chem. 1975, 53, 2975.
[16] W. Dullenkopf, J.-C. Castro-Palomino, L. Manzoni, R. R.
Schmidt, Carbohydr. Res. 1996, 296, 135.
The value of the synthesis of such a complex and sensitive
compound as 18 is not only that sufficient amounts of material
are provided for the investigation of its structure and for its
use in biological model studies; such a total synthesis also
discloses a strategy for the synthesis of structural analogues
that may have modified effects and potentially improved
biological stability. These options are generally not provided
by enzymatic syntheses. In the case of the PSGL-1binding site
glycopeptide 18, its use as a substrate for tyrosine O-sulfatyl
transferases and for the determination of the inhibitory
potential of both the unsulfated and the sulfated compounds
toward P-selectin will be investigated.
[17] H. Lönn, Carbohydr. Res. 1985, 139, 105.
[18] a) R. U. Lemieux, K. B. Hendriks, R. V. Stick, K. James, J. Am.
Chem. Soc. 1975, 97, 4056; b) S. Sato, M. Mori, Y. Ito, T. Ogawa,
Carbohydr. Res. 1986, 155, C6.
[19] M. Sakagami, H. Hamana, Tetrahedron Lett. 2000, 41, 5547.
[20] B. Liebe, H. Kunz, Angew. Chem. 1997, 109, 629; Angew. Chem.
Int. Ed. Engl. 1997, 36, 618.
Received: December 17, 2007
Published online: March 20, 2008
[21] A. Marra, P. Sinaþ, Carbohydr. Res. 1990, 195, 303.
[22] H. Kunz, C. Unverzagt, Angew. Chem. 1988, 100, 1763; Angew.
Chem. Int. Ed. 1988, 27, 1697.
Keywords: glycopeptides · glycosylation · solid-phase synthesis ·
T antigens · total synthesis
.
[23] 14: Yield: 165 mg (34% based on 13); colorless amorphous
solid; [a]2D3 = ꢀ16 degcm3 gꢀ1 dmꢀ1 (c = 1.0 gcmꢀ3, CHCl3); Rf =
0.54 (CH2Cl2/MeOH, 25:2), Rf = 0.19 (CH2Cl2/MeOH, 50:3);
Analyt. HPLC: Rt = 35 min (Phenomenex Luna C18, gradient:
CH3CN/H2O 5:95!100:0 in 40 min; l = 214 nm). 1H NMR [1H-
1H COSY, TOCSY] (400 MHz, CDCl3): d = 5.69 (d, 1H,
3JNH,Thra = 8.6 Hz, NH-Thr), 5.50 (mc, 1H, H8-Sia), 5.44 (dd,
[1] Reviews: a) M. Sperandio, FEBS J. 2006, 273, 4377; b) D.
Vestweber, J. E. Blanks, Physiol. Rev. 1999, 79, 1 81 ; c) F. M.
Unger, Adv. Carbohydr. Chem. Biochem. 2001, 57, 207.
[2] Review: S. Chen, M. Fukuda, Methods Enzymol. 2006, 416, 371.
[3] L. Descheny, M. E. Gainers, B. Walcheck, C. J. Dimitroff, J.
Invest. Dermatol. 2006, 126, 2065.
[4] M. Steegmaler, A. Levinovitz, S. Isenmann, E. Borges, M.
Lenter, H. P. Kocher, B. Kleuser, D. Vestweber, Nature 1995,
373, 615.
[5] a) Review: R. D. Cummings, Braz. J. Med. Biol. Res. 1999, 32,
519; b) D. Sako, K. M. Comess, K. M. Barone, R. T. Camphau-
sen, D. A. Cumming, G. D. Shaw, Cell 1995, 83, 323; c) T.
Pouyani, B. Seed, Cell 1995, 83, 333; d) F. Li, P. P. Wilkins, S.
Crawley, J. Weinstein, R. D. Cummings, R. P. McEver, J. Biol.
Chem. 1996, 271, 3255; e) R. Kumar, R. T. Camphausen, F. X.
Sullivan, D. A. Cumming, Blood 1996, 88, 3872; f) V. Ram-
achandran, M. U. Nollert, H. Qiu, W.-J. Liu, R. D. Cummings, C.
Zhu, R. P. McEver, Proc. Natl. Acad. Sci. USA 1999, 96, 13771;
g) W. S. Somers, J. Tang, G. D. Shaw, R. T. Camphausen, Cell
2000, 103, 467.
[6] a) K. M. Koeller, M. E. Smith, C.-H. Wong, J. Am. Chem. Soc.
2000, 122, 742; b) K.-T. Huang, B.-C. Wu, C.-C. Lin, S.-C. Luo, C.
Chen, C.-W. Wong, C.-C. Lin, Carbohydr. Res. 2006, 341, 2151.
[7] A. Leppänen, P. Mehta, Y. B. Ouyang, T. Ju, J. Helin, K. L.
Moore, I. van Die, W. M. Canfield, R. P. McEver, R. D. Cum-
mings, J. Biol. Chem. 1999, 274, 24838.
1H, 3JH7,H6 = 2.7 Hz, 3JH7,H8 = 9.7 Hz, H7-Sia), 5.36–5.34 (m, 2H,
3
H1-Fuc {5.36, d, 3JH1,H2 = 3.9 Hz}, H4-Gal’ {5.34, d, JH4,H3
=
3.1Hz}), 5.31 (d, 1H, 3JH4,H3 = 2.4 Hz, H4-Fuc), 5.25 (sb, 1 H,
H4-GalN), 5.20 (dd, 1H, 3JH3,H2 = 10.9 Hz, JH3,H4 = 3.2 Hz, H3-
3
Fuc), 5.12 (d, 1H, 3JNH,H5 = 10.2 Hz, NH-Sia), 5.06 (dd, 1H,
3JH2,H1 = 7.7 Hz, 3JH2,H3 = 10.5 Hz, H2-Gal’), 5.00–4.84 (m, 6H,
H2-Fuc {4.98}, H4-Gal {4.94}, H5-Fuc {4.93}, H3-Gal’ {4.92}, H2-
Gal {4.89}, H4-Sia {4.87}), 4.80 (d, 1H, 3JH1,H2 = 2.9 Hz, H1-
GalN), 4.70 (d, 1H, 3JH1,H2 = 8.0 Hz, H1-Gal), 4.63 (d, 1H,
2JH6a,H6b = 11.1 Hz, H6a-GlcN), 4.56–4.43 (m, 6H, H1-GlcN
{4.55}, H1-Gal’ {4.55}, H3-Gal {4.53}, CH2a-Fmoc {4.53}, H2-
GalN {4.48}, CH2b-Fmoc {4.45}), 4.36 (dd, 1H, 3JH6a,H5 = 6.7 Hz,
2JH6a,H6b = 11.5 Hz, H6a-Gal), 3.43 (mc, 1H, H6b-GalN), 2.57 (dd,
2
1H, JH3eq,H3ax = 12.6 Hz, 3JH3eq,H4 = 4.6 Hz, H3eq-Sia), 1.27 (d, 3H,
3JThrg,Thrb = 5.9 Hz, Thrg), 1.16 ppm (d, 3H, 3JH6abc,H5 = 6.2 Hz,
H6a,b,c-Fuc).
13C NMR
[BB,
HSQC,
HMBC]
(100.6 MHz,CDCl3): [d = 100.67 (C1-GlcN, C1-Gal’), 99.84
(C1-Gal), 99.25 (C1-GalN), 96.72 (C2-Sia), 95.22 (C1-Fuc),
83.08 (C(CH3)3), 75.60 (Thrb), 67.79 (C6-GalN), 58.92 (Thra),
18.58 (Thrg), 15.80 ppm (C6-Fuc). ESI-MS (positive): m/z =
1148.5 ([M+2Na]2+, calcd.: 1148.4), 2273.9 ([M+Na]+, calcd.:
2273.8).
[24] Rapp Polymere, Tübingen, Germany.
3448
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 3445 –3449

Angewandte
Chemie
3
[25] K. Barlos, D. Gatos, J. Kallitsis, G. Papaphotiu, P. Sotiriu, Y.
Wenqing, W. Schäfer, Tetrahedron Lett. 1989, 30, 3943.
[26] a) R. Knorr, A. Trzeciak, W. Bannwarth, D. Gillessen, Tetrahe-
dron Lett. 1989, 30. 1927; b) V. Dourtoglou, B. Gross, Synthesis
1984, 572.
8.8 Hz}, NH-Tyr1 {7.98, d, JNH,Tyr1a = 8.2 Hz}, NH-Leu2 {7.98, d,
3JNH,Leu2a = 8.2 Hz}, NH-Sia {7.94, d, 3JNH,H5 = 7.6 Hz}), 7.85 (d,
3
1H, JNH,Thra = 8.2 Hz, NH-Thr), 7.70–7.64 (m, 2H, NH-Tyr2
3
3
{7.69, d, JNH,Tyr2a = 7.6 Hz}, NH-Phe {7.65, d, JNH,Phea
=
7.2 Hz}), 4.80 (d, 1H, 3JH1,H2 = 2.9 Hz, H1-Fuc), 4.71–4.65 (m,
3
[27] a) L. A. Carpino, A. El-Faham, C. A. Minor, F. Albericio, J.
Chem. Soc. Chem. Commun. 1994, 201; L. A. Carpino, H.
Imazumi, A. El-Faham, F. J. Ferrer, C. Zhang, Y. Lee, B. H.
Foxman, P. Henklein, C. Hanay, C. Mügge, H. Wenschuh, J.
Klose, M. Beyermann, M. Bienert, Angew. Chem. 2002, 114, 457;
Angew. Chem. Int. Ed. 2002, 41, 441.
2H, H1-GalN {4.70, d, JH1,H2 = 3.3 Hz}, H5-Fuc {4.67}), 1.07 (d,
3H, JThrg,Thrb = 5.6 Hz, Thrg), 0.97 (d, 3H, 3JH6abc,H5 = 6.2 Hz,
3
d
abc
d
H6a,b,c-Fuc), 0.88–0.81ppm (m, 12H, Leu
, Leu2 abc).
1
100.6 MHz-13C NMR [BB, HSQC] (100.6 MHz, [D6]DMSO):
[d = 104.79 (C1-Gal’), 102.52 (C1-Gal), 100.92 (C1-GlcN), 99.06
(C1-Fuc), 98.98 (C1-GalN), 98.61 (C2-Sia), 76.00 (Thrb), 59.25
[28] 18: Yield: 31mg (64%); colorless lyophilisate; [ a]D23
=
(Pro1 ), 58.56 (Pro2 ), 56.09 (Thra), 55.21(C2-GlcN), 54.50
a
a
ꢀ23 degcm3 gꢀ1 dmꢀ1 (c = 0.96 gcmꢀ3, DMSO); Analyt. HPLC:
Rt = 13 min (Phenomenex Jupiter C18, gradient: CH3CN/H2O +
0.1% TFA 10:90 !100:0 in 40 min; l = 214 nm). 1H NMR [1H-
1H COSY, TOCSY] (400 MHz, [D6]DMSO): [d = 8.32 (db, 1H,
3JNH,Glu1a = 6.0 Hz, NH-Glu1), 8.23–8.19 (m, 2H, NH-Leu1 {8.22,
d, 3JNH,Leu1a = 8.4 Hz}, NH-Asp1 {8.20, d, 3JNH,Asp1a = 7.8 Hz}), 8.14
(Tyr1 ), 54.25 (Tyr2 ), 53.45 (Phea), 52.22 (C5-Sia), 51.78 (Glu1 /
a a a
Glu2 ), 51.05 (Leu2 ), 49.85, 49.77, 48.79 (Asp1a, Asp2 , Leu1 ,
a
a
a
a
Glu1 /Glu2a), 18.80 (Thrg), 16.64 ppm (C6-Fuc). HR-ESI-MS
a
(positive,
calcd.: 1356.5550).
+ ,
0.1% TFA): m/z = 1356.5509 ([M+1+2H]2+
[29] S. Keil, C. Claus, W. Dippold, H. Kunz, Angew. Chem. 2001, 113,
379; Angew. Chem. Int. Ed. 2001, 40, 366.
[30] R. Kuhn, I. Löw, Ber. Dtsch. Chem. Ges. 1941, 74, 219.
3
3
(d, 1H, JNH,Asp2a = 7.4 Hz, NH-Asp2), 8.07 (d, 1H, JNH,Glu2a
=
=
7.4 Hz, NH-Glu2), 8.03–7.93 (m, 4H, NH-GlcN {8.02, d, 3JNH,H2
Angew. Chem. Int. Ed. 2008, 47, 3445 –3449
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 3449