Trimethylsilylnitrate-Trimethylsilyl Azide: A Novel Reagent System for the Synthesis of 2-Deoxyglycosyl Azides from Glycals. Application in the Synthesis of 2-Deoxy-β-N-glycopeptides

Article abstract of DOI:10.1021/jo0354948

A novel reagent system comprising Me₃SiN₃ and 20 mol % of Me₃SiONO₂ permits conversion of glycals to 1-azido 2-deoxy sugars in one step in fair to good yields. Galactals offer higher stereoselectivities than do

Full text of DOI:10.1021/jo0354948

well. In this regard, Thiem et al.7 have recently reported
synthesis of a 2-deoxy analogue of nephritogenoside
noting the presence of 2-deoxysugars8 as part of several
naturally occurring antibiotics viz. anthracyclins. How-
ever, since most of the naturally occurring glycoproteins
have â-linkage of the peptide motif at the anomeric
center, it will also be of interest to develop methods for
the synthesis of â-linked 2-deoxy-N-glycopeptides.
Tr im eth ylsilyln itr a te-Tr im eth ylsilyl
Azid e: A Novel Rea gen t System for th e
Syn th esis of 2-Deoxyglycosyl Azid es fr om
Glyca ls. Ap p lica tion in th e Syn th esis of
2-Deoxy-â-N-glycop ep tid es
B. Gopal Reddy,† K. P. Madhusudanan,‡ and
Yashwant D. Vankar*,†
Clearly, one of the first steps toward the stereoselective
synthesis of N-linked glycopeptides requires methods to
stereoselectively introduce an amine functionality or its
equivalent at the anomeric center, which will eventually
lead to R- or â-linked N-glycopeptides. The scope of an
azido moiety as a good precursor of an amino group has
been well explored in carbohydrate chemistry and a
number of useful methods have been developed to
introduce it on carbohydrate molecules. Leaving groups
such as an acetate or a halide moiety (chloride or
bromide) at the anomeric center of a carbohydrate
molecule have been displaced by an azide group using
Me3SiN3 or metal azides (lithium, sodium, or silver) as
nucleophilic azide sources.9 This approach has been
applied10 in the synthesis of 2-deoxyglucofuranosyl azides.
More recently,11 an improved method to convert glycosyl
iodides into the corresponding glycosyl azides using
tetrabutylammonium azide or tetramethylguanidinium
azide has also been reported. Introduction of an azido
group on to a glycal derivative to obtain 2-deoxy-2-
halopyranosyl azides by reacting it with hypervalent
iodine reagents such as PhI(N3)2 or a combination of PhI-
(OAc)2 and Me3SiN3 has been reported by Kirschning et
al.12 Further, the Ferrier type of rearrangement, using
Department of Chemistry, Indian Institute of Technology
Kanpur, Kanpur-208 016, India, and Central Drug
Research Institute, Lucknow 226 001, India
vankar@iitk.ac.in
Received October 10, 2003
Abstr a ct: A novel reagent system comprising Me3SiN3 and
20 mol % of Me3SiONO2 permits conversion of glycals to
1-azido 2-deoxy sugars in one step in fair to good yields.
Galactals offer higher stereoselectivities than do the glucals.
Reduction of the azide group with Ph3P-H2O to amino
functionality followed by coupling with amino acids leads
to the synthesis of novel 2-deoxy-â-N-glycopeptides irrespec-
tive of the geometry of initial azido sugars. Using this
protocol, a new γ-sugar amino acid derivative is also
procured.
Glycopeptides play1,2 an important role in post-
translational biological selectivity such as cell growth
regulation, intracellular communication, cell adhesion,
cell differentiation, and many other cellular events.3 To
understand the precise role of various glycopeptides as
well as to modulate their activities, pure glycopeptides
need to be procured in sufficient quantity. As a result,
synthesis of naturally occurring O- and N-glycopeptides
as well as their modified analogues is an area of intense
study.4 Most of the glycoproteins have the peptide motifs
as â-linkage at the anomeric carbon, but isolation of
nephritogenoside,5 an R-linked N-glycopeptide, has
spurred6 some activity in developing methods for intro-
ducing R-linked amino group at the anomeric carbon as
Me3SiN3 along with a Lewis acid like Yb(OTf)3,13a Sc-
(OTf)3,13b and InBr3
has been employed to obtain
13c
enopyranosyl azides. In this paper, we wish to report a
new, one-step approach toward the synthesis of 2-deoxy-
1-azido sugars from glycals and their use in the synthesis
of 2-deoxy-â-N-glycopeptides.
We have been interested14 in developing newer meth-
odologies for functionalizing 2,3-glycals and their deriva-
tives en route to some useful carbohydrate synthons such
* To whom correspondence should be addressed. Fax: 0091-512-
259 0007.
(7) Laupichler, L.; Sowa, C. E.; Thiem, J . Biorg. Med. Chem. 1994,
2, 1281.
(8) (a) Thiem, J .; Klaffke, W. Top. in Curr. Chem. 1990, 154, 285.
(b) Roush, W. R.; Hartz, R. A.; Gustin, D. J . J . Am. Chem. Soc. 1999,
121, 1990.
(9) (a) Gyo¨rgydea´k, Z.; Szila´gyi, L.; Pausen, H. J . Carbohydr. Chem.
1993, 12, 139. (b) Gyo¨rgydea´k, Z.; Thiem, J . Carbohydr. Res. 1995,
268, 85.
(10) Sˇtimac, A.; Kobe, J . Carbohydr. Res. 2000, 329, 317.
(11) Ying, L.; Gervay-Hague, J . Carbohydr. Res. 2003, 338, 835.
(12) (a) Kirschning, A.; Hashem, M. A.; Monenschein, H.; Rose, L.;
Scho¨ning, K.-U. J . Org. Chem. 1999, 64, 6522. (b) Kirschning, A.;
J esberger, M.; Monenschein, H. Tetrahedron Lett. 1999, 40, 8999.
(13) (a) Kawabata, H.; Kubo, S.; Hayashi, M. Carbohydr. Res. 2001,
333, 153. (b) Yadav, J . S.; Reddy, B. V. S.; Chand, P. K. Tetrahedron
Lett. 2001, 42, 4057. (c) Yadav, J . S.; Reddy, B. V. S. Synthesis 2002,
511.
(14) (a) Rani, S.; Vankar, Y. D. Tetrahedron Lett. 2003, 44, 907. (b)
Pachamuthu, K.; Das, J .; Gupta, A.; Schmidt, R. R.; Vankar, Y. D. Eur.
J . Org. Chem. 2002, 1479. (c) Reddy, B. G.; Vankar, Y. D. Tetrahedron
Lett. 2003, 44, 4765. (d) Pachamuthu, K.; Vankar, Y. D. J . Org. Chem.
2001, 66, 7511. (e) Gupta, A.; Vankar, Y. D. Tetrahedron 2000, 56,
8525. (f) Rani, S.; Agarwal, A.; Vankar, Y. D. Tetrahedron Lett. 2003,
44, 5001.
† Indian Institute of Technology Kanpur.
‡ Central Drug Research Institute.
(1) (a) Varki, A. Glycobiology 1993, 3, 97. (b) Davis, B. G. J . Chem.
Soc., Perkin Trans. 1 1999, 3215. (c) Arsequell, G.; Valencia, G.
Tetrahedron: Asymmetry 1999, 10, 3045.
(2) (a) Bertozzi, C. R.; Kiessling, L. Science 2001, 291, 2357. (b)
Rudd, P. M.; Elliott, T.; Creswell, P.; Wilson, I. A.; Dwek, R. A. Science
2001, 291, 2370.
(3) Dwek, R. A.; Butters, T. D.; Platt, F. M.; Zitzmann, N. Nature
2002, 1, 65.
(4) (a) Davis, B. G. Chem. Rev. 2002, 102, 579. (b) Herzner, H.;
Reipen, T.; Schultz, M.; Kunz, H. Chem. Rev. 2000, 100, 4495. (c)
Dondoni, A.; Marra, A. Chem. Rev. 2000, 100, 4395. (d) Peluso, S.;
Imperiali, B. Tetrahedron Lett. 2001, 42, 2085. (e) Dondoni, A.;
Mariotti, G.; Marra, A. Tetrahedron Lett. 2000, 41, 3483. (f) Nolen, E.
G.; Kurish A. J .; Wong, K. A.; Orlando, M. D. Tetrahedron Lett. 2003,
44, 2449.
(5) (a) Shibata, S.; Miyagawa, Y.; Naruse, T.; Takuma, T. J .
Immunol. 1969, 102, 593. (b) Shibata. S.; Nagasawa, T. J . Immunol.
1971, 106, 1284.
(6) (a) Shibata, S.; Takeda, T.; Utsuno, A.; Okamoto, N.; Ogihara,
Y. Carbohydr. Res. 1990, 207, 71. (b) Sasaki, M.; Tachibana, K.;
Nakanishi, H. Tetrahedron Lett. 1991, 32, 6873.
10.1021/jo0354948 CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/06/2004
2630
J . Org. Chem. 2004, 69, 2630-2633

as 1,2-dihydroxy sugars,14a 2-acetoxy 1,2-glycals,14a
2-amino-â-C-glycosides,14b N-acetyl-R-D-lividosaminide,14c
2-deoxy-O-glycosides,14d C-2-methyl O- and C-glycosides
and sugar-derived R-methylene-δ-valerolactones.14e More
recently, we have found14f that a new reagent system viz.
LaCl3‚7H2O/NaI/PhCH2OH is suitable to hydrate glycals
and this methodology was used in synthesizing 1,6-
dideoxynojirimycin. A few years ago we developed15 a new
reagent system viz. Me3SiONO2-CrO3 to convert olefins
into R-nitroketones in one step. The putative intermedi-
ate was proposed to be Me3SiOCrO2ONO2, which acted
as a source of NO2+ and -OCrO3SiMe3 ions. We therefore
presumed that a combination of Me3SiONO2-Me3SiN3
could react with olefins, including glycals, to form the
corresponding 2-nitroazides with the loss of Me3SiOSiMe3.
It was expected that from the sugar derived 2-nitroazides
the NO2 group could be reductively removed to obtain
1-azido-2-deoxy sugars which can be further converted
to 2-deoxy-N-glycopeptides.
TABLE 1. Con ver sion of Glyca ls to 1-Azid o Su ga r s
In initial experiments, we found that cyclohexene and
styrene were unreactive toward this reagent system viz.
Me3SiONO2-Me3SiN3 (1:1) but 3,4-dihydro-2H-pyran
reacted with it to form tetrahydropyran-2-azide in 94%
yield. We, therefore, studied the reaction of 3,4,6-tri-O-
benzyl-D-glucal 1f (Table 1) with this reagent system and
obtained a 2:1 mixture of R and â anomers of 1-azido-2-
deoxy-3,4,6-tri-O-benzyl-D-glucose 4f in 75% yield. Ir-
respective of the course of the reaction, the product
isolated indicates that eventual addition of the elements
-
of HN3 viz. H+ and N3 ions to compound 1f had occurred.
Further, glucal 1f was found to be unreactive toward a
freshly made solution (1.2 M) of HN3 which indicates that
the elements of HN3 were added to it in an indirect
manner in its reaction with Me3SiONO2-Me3SiN3. To the
best of our knowledge, there is no report in the literature
on the addition of HN3 to olefins. Since the product
formed in this reaction did not clearly indicate a role of
Me3SiONO2, we carried out experiments with less than
stoichiometric amounts of Me3SiONO2. We subsequently
found that 20 mol % of Me3SiONO2 (as a solution in CH3-
CN) was sufficient to form the 2-deoxy-1-azido sugars
from glycals. In these experiments, we had not isolated
Me3SiONO2, instead its bulk solution was made16 from
ClSiMe3 and AgNO3 and 20 mol % was taken out for each
experiment. Galactal derivatives gave better selectivity
than the glucal derivatives which gave a mixture of the
R and â-glycosyl azides (entries 4f, 4g). The galactal
derivatives 1a , 1b, and 1e (Table 1) gave the correspond-
ing 2-deoxy-1-azido sugars 4a , 4b, and 4e having 4C1
conformation with the azide moiety in R-orientation.
However, 1c and 1d gave azido sugars 4c and 4d which
possessed 1C4 conformations.17 Our results are sum-
marized in Table 1. To the best of our knowledge, this is
the first report of converting glycals to 2-deoxy-1-azido
sugars in one step, although there are indirect two- or
three-step procedures to procure them from glycals via
HBr addition followed by further manipulations.18
a
b
Yields of chromatographically pure compounds. R/â ) 2:1.
c R/â ) 1.6:1.
Exposure of 3,4,6-tri-O-benzyl-D-glucal 1f (Scheme 1
where X ) H, Y ) OBn, R ) Bn) to Me3SiONO2 (1.1
molar equiv) followed by analysis of the reaction mixture
by thin-layer chromatography showed a polar spot com-
pared to the starting material. The reaction mixture was
then treated with Me3SiN3 followed by aqueous workup
to yield 1-azido-2-deoxy sugar 4f (Scheme 1). If, on the
other hand, the reaction was not treated with Me3SiN3
(17) We thank one of the reviewers, who suggested that we reconfirm
the configurations of 4c and 4d which, originally, we had perceived to
have 4C1 conformations. Accordingly, homonuclear decoupling experi-
ments were performed using the two methylene (H-2) protons in 4c.
Thus, irradiation of the signal for one of the two methylene protons at
C-2 (H-2e) at δ 2.17-2.23 led to the signal for H-1 to appear as a
doublet with J ) 8.08 Hz. On the other hand, H-1 appeared as a
doublet with J ) 5.8 Hz when irradiation of the other C-2 methylene
hydrogen (H-2a) at δ 1.58-1.65 was performed. This confirms that H-1
and H-2a are diaxially oriented in 4c and correspondingly the azide
group is equatorial. Further, in an NOE experiment, irradiation of the
signal for H-1 did not enhance the signals for H-3 and H-5; instead,
only the signal for H-2e at δ 2.17-2.23, as one would expect, was
enhanced. This suggests that H-1 and H-5 have a trans relationship
and hence this aspect coupled with the fact that H-1 is axially oriented
clearly confirms that 4c has 1C4 conformation. Likewise, the configu-
ration at H-1 in 4d and its structure was established.
(15) (a) Shahi, S. P.; Gupta, A.; Pitre, S. V.; Reddy, M. V. R.;
Kumareswaran, R and Vankar, Y. D. J . Org. Chem. 1999, 64, 4509.
(b) Reddy, M. V. R.; Kumareswaran, R.; Vankar, Y. D. Tetrahedron
Lett. 1995, 36, 7149.
(16) Kimura, M.; Kajita, K.; Onoda, N.; Morosawa, S. J . Org. Chem.
1990, 55, 4887.
(18) (a) Ollivier, C.; Renaud, P. J . Am. Chem. Soc. 2001, 123, 4717.
(b) Huang, X.; Surry, C.; Hiebert, T.; Bennet, A. J . J . Am. Chem. Soc.
1995, 117, 10614. (c) Szarek, W.; Achmatowicz, J r. O.; Plenkiewicz,
J .; Radatus, B. K. Tetrahedron 1978, 34, 1427.
J . Org. Chem, Vol. 69, No. 7, 2004 2631

SCHEME 1
derivative to form the corresponding glycosyl amides. In
the present work, however, we have used Ph3P along with
H2O23 to reduce the azide functionality and without
isolating the free amine it was coupled with an activated
amino acid under standard conditions. Interestingly,
irrespective of the geometry of the anomeric azide, the
geometry of the final coupled product was â at the
anomeric center in the four glycopeptides (5a -d , Scheme
2) that we have synthesized. Thus, the azido sugar 4f (a
mixture of R and â anomers), upon reduction followed
by coupling with a monopeptide viz. 1-succinimidyl-N-
tert-butyloxycarbonylglcylglycinate gave exclusively the
â-linked glycodipeptide 5d . It is, however, possible that
a small amount of the R-linked glycopeptide may have
formed but we could neither isolate the same nor could
trace its presence by the NMR spectroscopy of the product
isolated by column chromatography. Formation of â-ano-
meric 1-amino sugars in the cases studied is not unex-
pected since a somewhat similar observation was re-
ported24 recently while reducing 1-azidosugars using H2/
Lindlar catalyst where exclusively â-anomeric 1-amino
sugars are obtained irrespective of the geometry of the
starting anomeric azides. Similarly, NH4HCO3 is also
known25 to convert reducing sugars eventually into
â-anomeric 1-amino sugars. We believe that this one-step
methodology to obtain 2-deoxy-1-azido sugars followed
by their conversion to 2-deoxy-N-glycopeptides7,26 will
find wide application.
and solvent evaporated under vacuum then the product
obtained showed in its IR spectrum a strong peak at 1680
cm-1 whereas its 1H NMR spectrum showed peaks of
no specific splitting patterns. Mass spectral analysis
using electrospray technique (+ ion) suggested the pres-
ence of 1-hydroxy sugar derivative 3; however, the -ve
ion detection indicated the presence of a peak at m/z
479 corresponding to the nitrate ester 2. Clearly, at this
stage the product appears to be a mixture of several
components. It is therefore likely that the product of
the first reaction is compound 2 which is relatively
unstable. Further, if the reaction was first worked up
with water and the crude product treated with Me3SiN3
then the only product isolated after column chromatog-
raphy was 2-deoxy-3,4,6-tri-O-benzyl-D-glucose 3 (Scheme
1). This is not unexpected since it is known19 that ano-
meric nitrates are susceptible toward hydrolysis on
column chromatography. Further, during workup also
the nitrate ester 2 might get hydrolyzed to the hydroxy
sugar 3. We, therefore, presume that while preparing
Me3SiONO2 from Me3SiCl and AgNO3 in CH3CN suf-
ficient amount of H+ ions (possibly due to dissolved HCl
in ClSiMe3) must have been present20 in the solution
which allowed the addition of the elements of HNO3 to
a glycal to form glycosyl nitrate 2 which then sub-
sequently reacts with Me3SiN3 to yield the azide 4.
Since Me3SiONO2 is recovered back in the reaction of
nitrate ester 2 with Me3SiN3, only 20 mol % of it is
sufficient for the completion of the reaction as the
recycled Me3SiONO2 again reacts with a glycal to form
glycosyl nitrate 2.
We have further utilized the present methodology in
synthesizing a 2-deoxy sugar derived γ-amino acid ester
8 as shown in Scheme 3. Thus, 6-O-TBDMS-protected
3,4-di-O-benzyl-D-galactal 1b was converted into R-1-
azido-2-deoxy sugar 4b in 55% yield by using Me3-
SiONO2-Me3SiN3 reagent system. Reduction of the azide
4b with Ph3P-H2O gave the corresponding â-amino
sugar derivative whose acetylation gave N-acetylamino
sugar 6 in 70% yield. Deprotection of -OTBDMS group
using n-Bu4NF followed by oxidation with NaOCl-
TEMPO9b,11 gave the desired amino acid which was
characterized as the corresponding methyl ester 8. In
view of the importance27 of sugar amino acids in the
synthesis of modified peptidomimetics, we believe that
the sugar amino acid derivative 8 will be useful.
In summary, we have developed a new reagent system
to convert glycals into 1-azido-2-deoxy sugars. The utility
of these azido sugars in the synthesis of 2-deoxy-â-N-
glycopeptides and a sugar amino acid has been demon-
strated. We expect that the new reagent system Me3-
SiN3-Me3SiONO2, as well as the methodology of formation
of 2-deoxy-â-N-glycopeptides and the sugar-derived γ-ami-
no acid will find use in further studies.
To convert these 2-deoxy-1-azido sugars to the corre-
sponding glycopeptides, we studied their reduction fol-
lowed by coupling with appropriate amino acids. Reduc-
tion of anomeric azides has drawn a lot of attention in
recent years,21 and in this regard, the pioneering work
by DeShong22 has led to the stereoselective formation of
R-glycosyl amides via their isoxazoline derivatives by
neighboring group participation of the acetyl group at
C-2. In all these cases, the in situ generated intermedi-
ates were trapped by either a carboxylic acid or its
(23) Suga, H.; Ersoy, O.; Tsumuraya, T.; Lee, J .; Sinskey, A. J . J .
Am. Chem. Soc. 1994, 116, 487.
(24) Suzuki, T.; Suzuki, S. T.; Yamada, I.; Koashi Y.; Yamada, K.;
Chida, N. J . Org. Chem. 2002, 67, 2874 and references therein.
(25) Cohen-Anisfeld, S. T.; Lansbury, P. T. J . Am. Chem. Soc. 1993,
115, 10531.
(26) For other syntheses of 2-deoxy-N-glycoppetides, see: (a) Isac-
Garcia, J .; Carvo-Flores, F. G.; Hernandez-Mateo, F.; Santoyo-Gonza-
lez, F. Eur. J . Org. Chem. 2001, 383. (b) Urge, L.; Otvos, J r. L.; Lang,
E.; Wroblewski, K.; Laczko, I.; Hollosi, M. Carbohydr. Res. 1992, 235,
83.
(19) Lemieux, R. U.; Ratcliffe, R. M. Can. J . Chem. 1979, 57,
1244.
(20) The pH of the reaction mixture was found to be 2.5.
(21) (a) Boullanger, P.; Maunier, V.; Lafont, D. Carbohydr. Res.
″
2000, 324, 97. (b) Kova´cs, L.; Osz, E.; Domokos, V.; Holzer, W.;
Gyo¨rgydea´k, Z. Tetrahedron 2001, 57, 4609. (c) Sridhar, P. R.; Prabhu,
K. R.; Chandarsekaran, S. J . Org. Chem. 2003, 68, 5261 and references
therein.
(27) (a) Gruner, S. A. W.; Locardi, E.; Lohof, E.; Kessler, H. Chem.
Rev. 2002, 102, 491. (b) Schweizer, F. Angew. Chem. 2002, 114, 240.
Angew. Chem., Int. Ed. 2002, 41, 230.
(22) Damkaci, F.; DeShong, P. J . Am. Chem. Soc. 2003, 125,
4408.
2632 J . Org. Chem., Vol. 69, No. 7, 2004

SCHEME 2
SCHEME 3
Exp er im en ta l Section :
Ack n ow led gm en t. We thank the Department of
Science and Technology, New Delhi, for financial sup-
port (through Grant No. SP/S1/G-21/2001). We also
thank one of the reviewers for pointing out some of the
references that we had omitted.
Gen er a l P r oced u r e for th e Syn th esis of 2-Deoxy Gly-
cop yr a n osyl Azid e. To a stirred solution of Me3SiONO2
16
(∼20 mol %) in freshly distilled dry CH3CN (2 mL) at 0 °C
was added TMS-N3 (37 µL, 0.288 mmol). A solution of a glycal
(0.240 mmol) in CH3CN (2 mL) was then added drop-
wise to the reaction mixture. The cooling bath was removed
and stirring continued for a period as indicated in Table 1. The
reaction mixture was then extracted with diethyl ether (2
× 20 mL) followed by usual workup to give a residue which
was purified by column chromatography to give an azido
sugar.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and spectral data for compounds 4a -g, 5a -d , and
6-8. This material is available free of charge via the Internet
at http://pubs.acs.org.
J O0354948
J . Org. Chem, Vol. 69, No. 7, 2004 2633