New types of glycoconjugates: O-glycosylated, N-glycosylated and O-,N-diglycosylated isoserine derivatives

Article abstract of DOI:10.1016/S0040-4039(03)00896-7

Starting from hexafluoroacetone-protected malic acid O-glycosylated, N-glycosylated and O-,N-diglycosylated (S)-isoserine derivatives have been synthesized. The new compounds represent glycosylated β-alanine surrogates, i.a. suitable for β-peptide modific

Full text of DOI:10.1016/S0040-4039(03)00896-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 4223–4226
New types of glycoconjugates: O-glycosylated, N-glycosylated
and O-,N-diglycosylated isoserine derivativesꢀ
Christoph Bo¨ttcher and Klaus Burger*
Department of Organic Chemistry, University of Leipzig, Johannisallee 29, D-04103 Leipzig, Germany
Received 3 March 2003; accepted 4 April 2003
Abstract—Starting from hexafluoroacetone-protected malic acid O-glycosylated, N-glycosylated and O-,N-diglycosylated (S)-iso-
serine derivatives have been synthesized. The new compounds represent glycosylated b-alanine surrogates, i.a. suitable for
b-peptide modification. © 2003 Elsevier Science Ltd. All rights reserved.
Glycoproteins are ubiquitous in nature, playing a key
role in various biological recognition processes on
membranes.2,3 The peptide and carbohydrate part are
linked together by glycosidic bonds.4 There are two
modes for the attachment of glycosides to the peptide
backbone involving either the oxygen atom in the side
chain of serine and threonine, or the nitrogen atom in
the side chain of asparagine.5 These bonds are poten-
tially acid-sensitive and in some cases labile even under
moderate basic conditions.3 Glycopeptides contain
numerous functional groups. Therefore, the regio- and
stereoselective synthesis of glycopeptides requires
sophisticated, selective and mild protection and activa-
tion strategies.2b,4
of the amide group of N-glycosylated species by a
retroamide subunit10 or an ethylene isoster,11 and by
Imperiali et al. are the incorporation of alanine-b-
hydroxylamine9
and
alanine-b-hydrazide9,12
as
asparagine surrogates. Recently, a new approach to
glycopeptide mimetics was disclosed, where O- and
N-glycosidic linkages are replaced by ureaꢀglycosyl
bonds.13 The ureaꢀglycosyl bonds were constructed by
coupling glycosyl isocyanates and amino acid
derivatives.
Herein we report on the incorporation of isoserine
directly at the interface carbohydrate/peptide and on an
alternative synthetic approach for an efficient incorpo-
ration of an urea linkage between the carbohydrate
backbone and a b-amino acid using hexafluoroacetone
as protecting and activating reagent.
In general, glycoconjugates isolated from biological
sources or produced by gentechnological methods are
microheterogeneous and can not be applied for biologi-
cal studies. Therefore, the development of new method-
ology for the assembly of more robust, homogeneous
glycopeptide mimetics as well as the construction of
new binding motifs by chemoselective ligation is of
current interest, i.a. glycoconjugates have been synthe-
sized where O- and N-glycosidic linkages are replaced
by non-natural linkages, like carbonꢀcarbon,6,7 car-
bonꢀsulfur7,8 and carbonꢀaminooxy units.9 Alternative
concepts proposed by Kessler et al. are the replacement
[(5S)-2,2-Bis(trifluoromethyl)-4-oxo-1,3-dioxolan-5-yl]-
methyl isocyanate (2)14 readily obtained from malic
acid (1) and hexafluoroacetone is excellently suited for
regioselective trifunctionalization.14,15 Fluoren-9-yl-
methanol adds exclusively across the isocyanate group
of 2 to give the fully protected, a-carboxy-activated
Fmoc-(S)-isoserine derivative, which on treatment with
an excess of allyl alcohol reacts to give the correspond-
ing allyl ester 3 (Scheme 1). Concomitantly, the
hydroxy group is deblocked, and can be glycosylated in
the next step, for example applying Schmidt’s imidate
strategy.16 Compound 4 is an O-glycosylated orthogo-
nally protected b-amino a-hydroxy acid. Deblocking of
the carboxy moiety can be achieved by applying the
protocol of Kunz and Waldmann17 to obtain O-glyco-
sylated Fmoc-Ise-OH (5), a new building block for
glycopeptide modification.
Keywords: hexafluoroacetone; (S)-isoserine; b-amino acids; glycosyla-
tion; peptide modification.
ꢀ Neoglycoconjugates, Part 4. For Part 3, see Ref. 1.
* Corresponding author. Fax: (49) 341 9736599; e-mail: burger@
organik.chemie.uni-leipzig.de
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00896-7

4224
C. Bo¨ttcher, K. Burger / Tetrahedron Letters 44 (2003) 4223–4226
Scheme 1. Reagents and conditions: (i) (CF3)2CO, DMSO; (ii) SOCl2, reflux; (iii) TMSN3, toluene, 80°C; (iv) fluoren-9-yl-
methanol, CHCl3, reflux; (v) excess allyl alcohol, CHCl3, reflux, (vi) 1.2 equiv. a-
D
-Ac4-Glc{Gal}-O-C(ꢁNH)CCl3, cat. TMSOTf,
,
MS 4 A, CH2Cl2, (vii) (PPh3)4Pd, N-methylaniline, THF.
By a slightly modified procedure (Scheme 2) compound
2 can be transformed into O-glycosylated dipeptide
derivatives 7b,c. After introduction of the Fmoc group,
the cleavage of the lactone ring is now performed with
an a-amino acid ester (2“6). In a consecutive step the
hydroxy group is glycosylated on treatment with the
corresponding imidate-activated carbohydrate building
block. O-glycosylated dipeptides of type Fmoc-Ise-Xaa-
OMe (7b: Fmoc-Ise-Phe-OMe) are obtained in modest
yields.
Dipeptides of type Fmoc-Xaa-Ise-OMe (7c: Fmoc-Phe-
Ise-OMe) with O-glycosylated isoserine in C-terminal
position are obtainable from 2 in a three step sequence,
consisting of a Goldschmidt reaction,14 alcoholysis, and
finally O-glycosylation (Table 1).
Scheme 2. Reagents and conditions: (i) 1.2 equiv. a-
D
-Ac4-
,
Glc-O-C(ꢁNH)CCl3, cat. TMSOTf, MS 4 A.
Table 1.
No. R1
R2
Solvent
Yield (%)
N-Glycosylation with introduction of an urea linkage
can be achieved in a preparatively simple way on
reaction of 2 with the corresponding glycosylamine
(2“8) (Scheme 3). Compounds 8 are obtained as
anomeric mixtures, the b-anomers are the main prod-
ucts (Table 2). 8a-b, 8b-b, 8d-b have been isolated
analytically pure from the crude reaction mixtures by
flash chromatography. However, experiments to sepa-
rate 8c-a and 8c-b failed. Compounds 8 are a-carboxy-
activated species. Therefore, they can be directly
submitted to peptide synthesis. Dipeptide formation is
coupled with a simultaneous deprotection of the
hydroxyl group (8“9). Compounds 9 are N-glycosyl-
ated dipeptides with an unprotected a-hydroxy group
which can be functionalized further. Therefore, they
represent building blocks, i.e. suitable for the construc-
tion of glycosylated depsipeptides.
7a
7b
7c
Fmoc
Fmoc
Fmoc-Phe
OMe
Phe-OMe
OMe
CH2Cl2
CH2Cl2
CH2Cl2/THF 5:1
78
25
20
namely a 3-glycosylated (5S)-5-hydroxy-2,4-dioxo-5,6-
dihydropyrimidine (12) (Scheme 3). While conversion
of the glucose derivative proceeded with acceptable
yields (8a“12a), the mannose derivative 8c did no react
to give an analoguous cyclocondensation product under
identical reaction conditions. On prolonged heating 8c
decomposed slowly. An intramolecular ring closure of a
similar type was observed when N-glycosylated HFA-
protected malic acid derivatives were treated with pyri-
dine in chloroform under reflux.1
On heating a solution of 8 in chloroform in the pres-
ence of pyridine we expected an intramolecular cycliza-
tion to form a six-membered heterocyclic ring system,
When to a solution of 8-b in dichloromethane two
equivalents of TFA were added, a slow anomerization
process was induced (Scheme 4). Equilibrium ratios

C. Bo¨ttcher, K. Burger / Tetrahedron Letters 44 (2003) 4223–4226
4225
Scheme 3. Reagents and conditions: (i) 0.95 equiv. glycosylamine, CH2Cl2, 0°C; (ii) 1.2 equiv. HCl*H-Phe-OtBu, NEM, DMF; (iii)
25 equiv. pyridine, CHCl3, reflux.
Table 2.
No.
R1
R2
Pg
Yield (%) 1“8-b+a
Anomeric ratioa 8-b/8-a
Yield (%) 8-b“9
a
b
c
OAc
OBzl
H
H
H
OAc
OBzl
Ac
Bzl
Ac
Bzl
64
62
64
70
12:1
3:1
12:1
10:1
86
74
83
73
d
H
a Anomeric ratio of the crude product was determined by 1H and 19F NMR spectroscopy.
Scheme 4. Reagents and conditions: (i) 2 equiv. TFA, CH2Cl2.
1
were taken from the H and 19F NMR spectra after 36
h (Table 3).
arsenal of new types of oligosaccharides and glycocon-
jugates has to be developed.20
Another new type of building block, namely N-,O-
diglycosylated dipeptide allyl ester (11), was obtained
on O-glycosylation of 10 (Scheme 5). Since cleavage of
the lactone ring can be also performed with b-amino
acid esters diglycosylated b-dipeptides become accessi-
ble via this reaction sequence.
Table 3.
b
a
No.
Ratio 8-b/8-aa
J12 (Hz)b
J12 (Hz)b
8a
8b
8c
8d
4:1
1:3
4:1
3:2
9.6
9.0
ꢀ1
ꢀ1
4.8
2.5
–
Multivalency of the interactions between carbohydrates
and carbohydrate-binding proteins18 or nucleic acids19
has recently become a subject of considerable interest.
To study the factors governing the strength of these
cluster effects, methodology for the synthesis of an
3.3
a Anomeric ratio of the equilibrated mixture was determined by 1H
and 19F NMR spectroscopy.
b Coupling constants were measured in CDCl3.

4226
C. Bo¨ttcher, K. Burger / Tetrahedron Letters 44 (2003) 4223–4226
Scheme 5. Reagents and conditions: (i) excess allyl alcohol, CHCl3, reflux; (ii) 1.2 equiv. a-
D
-Ac4-Glc-O-C(ꢁNH)CCl3, cat.
,
TMSOTf, MS 4 A, CH2Cl2.
On O-,N-glycosylation and O-,N-diglycosylation of
multifunctional b-amino acids (a-methylisoserine, iso-
cysteine) and multifunctional g-amino acids (homoiso-
serine, a-methylhomoisoserine, homoisocysteine) apply-
ing the ‘hexafluoroacetone route’ we report elsewhere.
7. For reviews, see: (a) Schweizer, F. Angew. Chem., Int.
Ed. 2002, 41, 230–253 and references cited therein; (b)
Davis, B. G. Chem. Rev. 2002, 102, 579–601.
8. Ka¨sbeck, L.; Kessler, H. Liebigs Ann. Chem. 1997, 1,
165–167.
9. Peluso, S.; Imperiali, B. Tetrahedron Lett. 2001, 42,
2085–2087.
10. Hoffmann, M.; Burkhart, F.; Hessler, G.; Kessler, H.
Helv. Chim. Acta 1996, 79, 1519–1532.
11. Burkhart, F.; Hoffmann, H.; Kessler, H. Angew. Chem.,
Int. Ed. 1997, 36, 1191–1192.
Acknowledgements
We thank DFG (Bu 277/21-2) and FCI for financial
support.
12. Rodriguez, E. C.; Marcaurelle, L. A.; Bertozzi, C. R. J.
J. Org. Chem. 1998, 63, 7134–7135.
13. (a) Ichikawa, Y.; Nishiyama, T.; Isobe, M. Synlett 2000,
1253–1256; (b) Ichikawa, Y.; Nishiyama, T.; Isobe, M.
J. Org. Chem. 2001, 66, 4200–4205.
14. Burger, K.; Windeisen, E.; Pires, R. J. Org. Chem. 1995,
60, 7641–7645.
15. Ru¨hl, T.; Hennig, L.; Hatanaka, Y.; Burger, K.; Welzel,
P. Tetrahedron Lett. 2000, 41, 4555–4558.
16. Schmidt, R. R. Angew. Chem., Int. Ed. 1986, 25, 212–
235.
17. (a) Waldmann, H.; Kunz, H. Liebigs Ann. Chem. 1983,
1712–1725; (b) Kunz, H.; Waldmann, H. Angew. Chem.,
Int. Ed. 1984, 23, 71–72.
18. (a) Lee, Y. C.; Lee, R. T. Acc. Chem. Res. 1995, 28,
321–327; (b) Whitesides, G. M.; Mammen, M.; Choi, S.
Angew. Chem., Int. Ed. 1998, 37, 2754–2794.
19. Sreedhara, A.; Cowan, J. A. J. Biol. Inorg. Chem. 2001,
6, 166–172.
References
1. Bo¨ttcher, C.; Burger, K. Tetrahedron Lett. 2002, 43,
9711–9714.
2. (a) Varki, A. Glycobiology 1993, 3, 97–130 and refer-
ences cited therein; (b) Davis, B. G. Chem. Rev. 2002,
102, 579–601 and references cited therein.
3. Kunz, H. In Antibiotics and Antiviral Compounds;
Krohn, K.; Kirst, H. A.; Maag, H., Eds. Synthesis of
Biologically Interesting Glycopeptides; A Problem of
Chemical Selectivity, VCH, 1993; pp. 251–262 and refer-
ences cited therein.
4. Kunz, H. Angew. Chem., Int. Ed. 1987, 26, 294–308 and
references cited therein.
5. Dwek, R. A. Chem. Rev. 1996, 96, 683–720 and refer-
ences cited therein.
6. Lohof, E.; Planker, E.; Mang, C.; Burkhart, F.;
Dechantsreiter, M. A.; Haubner, R.; Wester, H.-J.;
Schwaiger, M.; Ho¨lzemann, G.; Goodman, S. L.;
Kessler, H. Angew. Chem., Int. Ed. 2000, 39, 2761–2764.
20. Katajisto, J.; Karskela, T.; Heinonen, P.; Lo¨nnberg, H.
J. Org. Chem. 2002, 67, 7995–8001.