Stereoselective approach to C-glycosylasparagines

Article abstract of DOI:10.1016/j.tetlet.2003.09.212

In this paper we describe a simple and efficient approach to N-glycopeptide analogues that incorporate a ketomethylene unit in place of a native amide link. Key C-C bond forming steps involve the stereocontrolled addition of a functionalised allylsilane to activated sugar derivatives and the asymmetric phase-transfer alkylation of a glycine imine. Utility of this chemistry is demonstrated by the synthesis of a C-analogue of the glycopeptide core found in nephritogenoside. Regioselective ozonolysis of a 1,5-diene is also described.

Full text of DOI:10.1016/j.tetlet.2003.09.212

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 9039–9041
Stereoselective approach to C-glycosylasparagines
Barry Lygo,* Benjamin I. Andrews and Daniel Slack
School of Chemistry, University of Nottingham, Nottingham NG7 2RD, UK
Received 13 August 2003; revised 18 September 2003; accepted 26 September 2003
Abstract—In this paper we describe a simple and efficient approach to N-glycopeptide analogues that incorporate a ketomethylene
unit in place of a native amide link. Key CꢀC bond forming steps involve the stereocontrolled addition of a functionalised
allylsilane to activated sugar derivatives and the asymmetric phase-transfer alkylation of a glycine imine. Utility of this chemistry
is demonstrated by the synthesis of a C-analogue of the glycopeptide core found in nephritogenoside. Regioselective ozonolysis
of a 1,5-diene is also described.
© 2003 Elsevier Ltd. All rights reserved.
The development of methods for the generation of
synthetic constructs which incorporate both amino acid
and carbohydrate elements is attracting increasing
attention.1 This is in part due to their potential as
mimics of natural glycopeptides2 and also because the
functionality resident in these materials is well-suited to
automated combinatorial diversification.3 For some
time we have been interested in the development of
synthetic approaches towards both carbohydrate4 and
peptide5 fragments in which heteroatoms found in the
native structures have been replaced by carbon. In this
paper we outline the development of a simple stereose-
lective route to N-glycosylasparagine-like structures
which incorporate an all-carbon connection between
the amino acid and carbohydrate units.
N-Glycopeptides are the most common type of gly-
copeptide found in nature and it has been estimated
that almost half of all natural proteins may contain this
structural element.6 The most common N-glycopeptide
link found involves the amino acid asparagine con-
nected to a polysaccharide via the amide side-chain. In
most cases these glycopeptides contain a common core
in which the first sugar unit attached is a b-linked
N-acetyl glucosamine (i.e. 1),2,7 however an increasing
number of N-glycosylasparagines involving alternative
carbohydrate attachments have been isolated and char-
acterised.1b Perhaps most notable amongst this latter
group is nephritogenoside 2, an N-glycopeptide isolated
from the glomerular basement membrane of rats.8 This
glycopeptide has been shown to induce glomeru-
lonephritis and this activity coupled with its unusual
structure has resulted in the development of a number
of synthetic approaches to this natural product.9
It has long been recognised that the replacement of
peptide bonds with their ketomethylene isosteres can
lead to bio-active compounds with improved metabolic
stability.10,11 Replacement of the amide link in N-glyco-
sylasparagines by this type of isostere (e.g. 3) appears
equally attractive as this minimal modification changes
the chemistry of both amide and the glycosidic bonds,
and hence should result in materials that are resistant
to hydrolysis at both these sites. However, although a
number of methods have been developed for the syn-
thesis of C-glycosylasparagines, to date, work in this
area has focused primarily on alternative amide
isosteres.2,12
In an effort to develop a general synthetic approach to
ketomethylene C-glycosylasparagines of type 3, we con-
* Corresponding author.
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.09.212

9040
B. Lygo et al. / Tetrahedron Letters 44 (2003) 9039–9041
sidered the possibility of using a bifunctional fragment
6 as a means of connecting a sugar derivative 4 to a
glycine derivative 5. This seemed like an attractive
approach to targets of this type as there is precedent for
good stereocontrol in the two carbonꢀcarbon bond
forming steps involved, and because the chemistry
required should be compatible with a range of different
sugar substrates 4.
dihydrocinchonidinium bromide as the phase-transfer
catalyst as we have previously established that this
quaternary ammonium salt generally gives high selec-
tivity for (S)-amino acid derivatives in reactions of this
type.15 Hydrolysis of the resulting imine, N-protection
with either Boc or Z, and oxidative cleavage of the
alkene then provided the target C-glycoside 9 in good
overall yield.
Thus this sequence of reactions constitutes a short (6
step) highly stereoselective route to the fully-protected
C-analogue of the glycopeptide core found in
nephritogenoside 2. The chemistry involved should be
compatible with a range of glycosyl fluorides and hence
should allow access to a range of novel C-linked gly-
copeptide structures.
We considered that the best strategy for connecting
these three fragments would involve initial coupling of
a sugar derivative 4 with allylsilane 6, followed by
alkylation with the glycine imine 5. In order to test this
hypothesis we first investigated the synthesis of 9, a
protected C-analogue of the glycopeptide core found in
nephritogenoside 2 (Scheme 1).
In an effort to probe the scope of this chemistry further
we have also investigated synthesis of C-linked gly-
copeptides of type 10.
The addition of simple allylsilanes to sugar derived
oxonium ions is known to proceed via axial attack, and
generally delivers good stereoselectivity for the a-C-gly-
coside with
D-glucose derivatives. After surveying a
These structures lack the C-2% and C-3% substituents
found in native N-glycopeptides, but retain the key C-4%
and C-6% hydroxyl functions, the positions at which
additional sugar units are usually attached.2 We believe
that compounds of this type have significant potential
as simplified glycopeptide mimics as they contain all the
functionality required for incorporation into more com-
plex glycopeptide fragments but avoid problems associ-
ated with the need to differentiate between secondary
hydroxyl groups. We envisaged that compounds of this
type should be accessible from the commercially-avail-
able tri-O-acetyl glucals 11a and 11b.
range of C-1 activation methods we established that, in
our hands, glycosyl fluoride 7 provided the most effec-
tive glucosyl donor for reaction with allylsilanes of type
6.13 Using this approach, reaction of 7 with commer-
cially-available 2-chloromethyl-3-trimethylsilylpropene
provided the C-glycoside 8 in good yield and with high
selectivity for the desired a-isomer. Chloride 8 was then
converted into the corresponding iodide which proved
to be an excellent substrate for the asymmetric PTC
alkylation using glycine imine 5.14 For this latter pro-
cess we employed O-benzyl-N-(9-anthracenylmethyl)-
It was found that glucals 11a and 11b could be coupled
with 2-chloromethyl-3-trimethylsilylpropene in excellent
yield using Yb(OTf)3 catalysis.16 As expected, good
selectivity for the a-product 12 was obtained in both
cases. Conversion to the corresponding iodides fol-
lowed by asymmetric phase-transfer alkylation, hydroly-
sis and Boc-protection gave intermediates 13 in good
overall yields (Scheme 2).14
It is interesting to note that although the phase-transfer
alkylation step required the use of 9M aqueous potas-
sium hydroxide, no hydrolysis of the O-acetyl groups
could be detected. Presumably this is because the phase-
transfer catalysts employed do not extract hydroxide
ion into the organic phase to any significant extent,17
and thus the two-phase nature of the reaction enables
even primary acetates to survive.
Scheme 1. Reagents and conditions: (i) 2-chloromethyl-3-
trimethylsilylpropene (1.5 equiv.), BF3·OEt2 (1.1 equiv.),
CH3CN, −30°C (64%); (ii) NaI, acetone; (iii) 5, O-benzyl-N-
(9-anthracenylmethyl)dihydrocinchonidinium bromide (10
mol%), 9 M aq. KOH, PhMe, rt; (iv) 15% aq. citric acid,
THF, rt; (v) BnOCOCl, Et3N, CH2Cl2, 0°C–rt (58% overall),
or Boc2O, Et3N, CH2Cl2, 0°C–rt (50% overall); (vi) O3,
CDCl3, −50°C; Ph3P, rt (93% R=Z, 94% R=Boc).
Next we required oxidative cleavage of the 1,1-disubsti-
tuted alkene in the presence of a 1,2-disubstituted
alkene. It was anticipated that this should be possible

B. Lygo et al. / Tetrahedron Letters 44 (2003) 9039–9041
9041
References
1. (a) Davis, B. G. Chem. Rev. 2002, 102, 579–601; (b)
Arsequell, G.; Valencia, G. Tetrahedron: Asymmetry
1999, 10, 3045–3094.
2. Dondoni, A.; Marra, A. Chem. Rev. 2000, 100, 4395–
4421.
3. Schweizer, F. Angew. Chem., Int. Ed. 2002, 41, 230–253.
4. (a) Lygo, B.; Rudd, C. N. Tetrahedron Lett. 1995, 36,
3577–3580; (b) Lygo, B. Synlett 1992, 793–795.
5. (a) Lygo, B.; Swiatyj, M.; Trabsa, H.; Voyle, M. Tetra-
hedron Lett. 1994, 35, 4197–4200; (b) Aloui, M.; Lygo,
B.; Trabsa, H. Synlett 1994, 115–116.
6. Apweiler, R.; Hermjakob, H.; Sharon, N. Biochem. Bio-
phys. Acta 1999, 1473, 4–8.
7. Roth, J. Chem. Rev. 2002, 102, 285–303.
8. Takeda, T.; Sawaki, M.; Ogihara, Y.; Shibata, S. Chem.
Pharm. Bull. 1989, 37, 54–56.
9. See for example: (a) Damkaci, F.; DeShong, P. J. Am.
Chem. Soc. 2003, 125, 4408–4409; (b) Zhang, H.; Wang,
Y.; Thurmer, R.; Al-Qawasmeh, R. A.; Voelter, W. Pol.
J. Chem. 1999, 73, 101–115; (c) Mizuno, M.; Muramoto,
I.; Kobayashi, K.; Yaginuma, H.; Inazu, T. Synthesis
1999, 162–165; (d) Takeda, T.; Kojima, K.; Ogihara, Y.
Carbohydr. Res. 1993, 243, 79–89; (e) Teshima, T.; Naka-
jima, K.; Takahashi, M.; Shiba, T. Tetrahedron Lett.
1992 33, 363–366.
10. See: Spatola, A. In Chemistry and Biochemistry of Amino
Acids, Peptides, and Proteins’; Weinstein, B., Ed.; Marcel
Dekker: New York, 1983; Vol. 7, pp. 267–357.
11. For examples of bioactive peptides incorporating
ketomethylene isosteres, see: (a) Steinmetzer, T.; Zhu, B.
Y.; Konishi, Y. J. Med. Chem. 1999, 42, 3109–3115; (b)
Dragovich, P. S.; Prins, T. J.; Zhou, R.; Fuhrman, S. A.;
Patick, A. K.; Matthews, D. A.; Ford, C. E.; Meador, J.
W.; Ferre, R. A.; Worland, S. T. J. Med. Chem. 1999, 42,
1203–1212 and references cited therein.
Scheme 2. Reagents and conditions: (i) 2-chloromethyl-3-
trimethylsilylpropene (1.2 equiv.), Yb(OTf)3 (10 mol%),
CH2Cl2, rt (12a, 96%, 12b, 87%); (ii) NaI, acetone; (iii) 5,
O-benzyl-N-(9-anthracenylmethyl)dihydrocinchonidinium bro-
mide (10 mol%), 9 M aq. KOH, PhMe, rt; (iv) 15% aq. citric
acid, THF, rt; (v) Boc2O, Et3N, CH2Cl2, 0°C–rt (13a, 55%
overall, 13b, 71% overall); (vi) O3, CDCl3, −50°C; Ph3P, rt (14a,
70%, 14b, 84%); (vii) H2, 10% Pd/C, EtOAc, rt (15a, 77%, 15b,
77%).
via ozonolysis since reaction at the 1,2-disubstituted
alkene should be disfavoured on both steric and electronic
grounds. This did indeed prove to be the case, however
low temperature and carefully-controlled delivery of
ozone were necessary in order to obtain good yields of
the desired products 14. Finally, hydrogenation of the
remaining alkene furnished the target compounds 15a and
15b.
12. For examples of synthetic approaches to N-glycosyl-
asparagines that do allow direct access to ketomethylene
isosteres, see: (a) Westermann, B.; Walter, A.; Flo¨rke, U.;
Altenbach, H.-J. Org. Lett. 2001, 3, 1375–1378; (b)
Werner, R. M.; Williams, L. M.; Davis, J. T.; Tetra-
hedron Lett. 1998, 39, 9135–9138.
This latter sequence of reactions helps to establish the
generality of this approach to C-glycosylasparagines and
demonstrates that good stereoselectivity can be obtained
with a number of different substrates.
13. Jegou, A.; Pacheco, C.; Veyrieres, A. Tetrahedron 1998,
54, 14779–14790.
14. Control experiments employing both achiral and (R)-
selective PTC’s established that the asymmetric alkyla-
tions outlined here proceed with ]90% d.e.
15. (a) Lygo, B.; Andrews, B. I. Tetrahedron Lett. 2003, 44,
4499–4502; (b) Lygo, B.; Andrews, B. I.; Crosby, J.;
Peterson, J. A. Tetrahedron Lett. 2002, 43, 8015–8018; (c)
Lygo, B.; Humphreys, L. D. Tetrahedron Lett. 2002, 43,
6677–6679; (d) Lygo, B.; Crosby, J.; Lowdon, T. R.;
Peterson, J. A.; Wainwright, P. G. Tetrahedron 2001, 57,
2403–2409; (e) Lygo, B.; Crosby, J.; Lowdon, T. R.;
Wainwright, P. G. Tetrahedron 2001, 57, 2391–2402; (f)
Lygo, B.; Crosby, J.; Peterson, J. A. Tetrahedron Lett.
1999, 40, 8671–8674; (g) Lygo, B.; Wainwright, P. G.
Tetrahedron Lett. 1997, 38, 8595–8598.
For all the substrates studied, the stereochemistry of the
C-glycoside bond is readily confirmed by 1H NMR
analysis. ThestereochemistryassignedtoC-2oftheamino
acid moiety is based on the well-established stereochem-
ical course of glycine imine alkylations involving O-benz-
yl-N-(9-anthracenylmethyl)dihydrocinchonidinium bro-
mide.14,15
In conclusion, we have developed a simple, stereoselective
approach to ketomethylene isosteres of N-glycosyl-
asparagines. This chemistry should allow access to a wide
range of novel glycopeptide-based materials and further
developments in this area will be reported in due course.
Acknowledgements
16. Takhi, M.; Abdel Rahman, A. A.-H.; Schmidt, R. R.
Tetrahedron Lett. 2001, 42, 4053–4056.
17. Lygo, B. Phase-Transfer Reactions, In Rodd’s Chemistry
of Carbon Compounds, Vol. V; Asymmetric Catalysis;
Elsevier Science Ltd.: Oxford, 2001; pp. 101–150.
We thank the EPSRC for funding. We would also like
to acknowledge use of the EPSRC’s Chemical Database
Service at Daresbury.