3-Azido-3-deoxy-glycopyranoside derivatives as scaffolds for the synthesis of carbohydrate-based universal pharmacophore mapping libraries

Article abstract of DOI:10.1016/S0960-894X(03)00384-6

The regiospecific syntheses of six monosaccharide scaffolds, 1-6, containing a carboxylic acid, an azido and a free hydroxyl group were accomplished through the utilization of a key intermediate, namely, methyl 3-azido-3-deoxy-β-D-glucopyranoside (10). Scaffold 2 was also used in generating combinatorial libraries using solid-phase methodologies.

Full text of DOI:10.1016/S0960-894X(03)00384-6

Bioorganic & Medicinal Chemistry Letters 13 (2003) 2185–2189
3-Azido-3-deoxy-glycopyranoside Derivatives as Scaffolds for the
Synthesis of Carbohydrate-Based Universal Pharmacophore
Mapping Libraries
Rakesh Jain,y Muthoni Kamau,{ Chunguang Wang, Robert Ippolito,x Huiming Wang,
Richard Dulina, Jan Anderson, David Gange{ and Michael J.Sofia *
Intercardia Research Labs, 8 Cedar Brooke Drive, Cranbury, NJ 08512, USAk
Received 21 January 2003; accepted 24 March 2003
Abstract—The regiospecific syntheses of six monosaccharide scaffolds, 1–6, containing a carboxylic acid, an azido and a free
hydroxyl group were accomplished through the utilization of a key intermediate, namely, methyl 3-azido-3-deoxy-b-d-glucopyr-
anoside (10).Scaffold 2 was also used in generating combinatorial libraries using solid-phase methodologies.
# 2003 Elsevier Science Ltd.All rights reserved.
One aspect of combinatorial chemistry focuses on gen-
erating chemically diverse libraries suitable for broad
screening.Our search for novel combinatorial platforms
for preparing broad screening libraries lead us to inves-
tigate monosaccharides as versitile scaffolds.Con-
ceptually, the use of a monosaccharide nucleus as a
scaffold to display substituents in a stereochemically
defined way was first pioneered by Hirschmann and co-
workers in their effort to identify a mimetic of a known
biologically active agent.1,2
hydroxyl groups on any natural sugar nucleus presents
a problem if one wants to use a sugar nucleus as a
combinatorial scaffold.3
In designing a practical monosaccharide scaffold system
we focused on building a system that would provide a
classical three-point pharmacophoric recognition motif
(Fig.1 ).4,5 Then to circumvent the chemoselectivity
problem, we used an easily differentiated functional
group triad that included a carboxylic acid, a free
hydroxyl group and a masked amino group.The two
remaining hydroxyl groups on a hexose nucleus would
be converted to their methyl ethers.Methyl ethers were
chosen because they provide permanent protecting
groups that are stable to a wide variety of chemical
transformations and we hoped that they would mini-
mize the contribution of these sites to molecular recog-
nition events maximizing the effect of the three point
motif.To accommodate a carboxylic acid at secondary
carbon sites, we chose to use a carboxymethyl unit.
Although not isosteric with a C-6 uronic acid, the car-
boxymethyl group could be easily introduced by direct
alkylation of a sugar hydroxyl group.
We were particularly attracted to monosaccharides as
combinatorial scaffolds because of the unique set of
characteristics they possess.The monosaccharide system
provides a rigid ring system with a defined three
dimensional spatial arrangement of substituents.The
high functional density of this nucleus provides multiple
sites for chemical modification.However, the difficulty
faced in trying to chemically differentiate each of the
*Corresponding author at current address: Bristol-Myers Squibb, 5
Research Parkway, Wallingford, CT 06492, USA.Fax: +1-203-677-
6984; e-mail: michael.sofia@bms.com
yCurrent address: Versicor Inc., 34790 Ardentech Court, Fremont, CA
94555, USA.
In our earlier report, our scaffolds incorporated a fluor-
enylmethoxycarbonyl (Fmoc) protected amino group as
the masked amino functionality.4,5 However, it became
apparent to us that the Fmoc group lacked the chemical
stability we required for doing a wide variety of chemical
transformations.To circumvent this problem, we chose
to replace the NHFmoc group with an azido group.
{Current address: Bristol-Myers Squibb, 206 and Provinceline Rd,
Princeton, NJ 08543, USA.
xCurrent address: ChemRoutes Corp., 3912-116th St., Edmonton,
Alberta, Canada T6J 1R4.
{Current address: Row2 Technologies, Inc.200 Lanidex Plaza, Par-
sippany, NJ 07054, USA.
kOperations ceased September 2000.
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd.All rights reserved.
doi:10.1016/S0960-894X(03)00384-6

2186
R. Jain et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2185–2189
Figure 1.
The scaffold design described above provides the
potential for accessing a large diverse set of scaffolds
that could be used to explore a wide range of three-
dimensional space.We envisioned that families of these
scaffolds, would find utility in mapping pharmacophore
recognition sites on target biomolecules and hence we
refer to these libraries a carbohydrate-based universal
pharmcophore mapping libraries.In this report we pre-
sent the synthesis of a family of 3-azido containing
monosaccharide scaffolds 1, 2, 3, 4, 5, and 6 and
demonstrate the construction of a combinatorial library
using scaffold 2.
other functionality (Scheme 2).This was accomplished
by first protecting the C-4 and C-6 hydroxyl groups as a
4,6-O-p-methoxybenzylidene cyclic acetal using anis-
aldehyde dimethyl acetal and p-toluenesulfonic acid
monohydrate in DMF to give 11.Acetylation of the free
hydroxyl group followed by reductive ring opening8
gave 13 in 96% yield.
With 13 providing the needed azido sugar with each of
the hydroxyl groups differentiated we were able to pro-
ceed with completing the synthesis of scaffold 1.This
was done by first methylating the C-4 hydroxyl group of
13 and removing the 6-O-p-methoxybenzyl group with
DDQ.9 Oxidation with Jones reagent10 followed by de-
O-acetylation produced compound 1 in 88% yield after
silica gel column chromatography.
The syntheses of the 3-azido scaffolds 1, 2, 3, 4, 5, and 6
began with construction of the common intermediate
methyl 3-azido-3-deoxy-b-d-glucopyranoside 10.Inter-
mediate 10 would allow us to construct scaffold systems
having a carboxylic acid unit or a free hydroxyl group in
the 2-, 4- or 6-position of a glucose or galactose nucleus.
For the synthesis of 10, we employed 1,2,4,6-tetra-O-
acetyl-3-azido-3-deoxy-d-glucopyranose (7) which was
prepared from 1,2-5,6-di-O-isopropylidene-a-d-allofur-
anose using a previously described procedure for the
preparation of 3-azido-3-deoxy-d-mannose.6 Treatment
Scaffolds containing
a
carboxylate functionality
attached to either the 2- or 4-position of the sugar
nucleus required that we introduce a carboxymethyl
substituent.In each case, we were able to accomplish
the attachment of this carboxymethyl group by alkyl-
ation of a free hydroxyl group with methyl 2-bromo-
acetate and Ag2O-KI.For the synthesis of scaffold
2
7
of 7 with Me3SiBr and BiBr3 in CH2Cl2 gave crude
containing a carboxymethyl group at C-2, we began
with compound 11 having a free hydroxyl at C-2.
Alkylation with methyl 2-bromo-acetate gave 17 in 80%
yield (Scheme 3).Conversion of the 4,6-benzylidene
derivative 17 to the 4-O-methyl-6-hydroxy derivative 20
was accomplished as previously described for the pre-
paration of compound 15.Hydrolysis of methyl ester in
compound 20 provide scaffold 2 in 94% yield.
bromide 8.Treatment of 8 with MeOH–CH2Cl2 in the
presence of Ag2O followed by de-O-acetylation with
methanolic sodium methoxide afforded 10 in 87% yield
after chromatographic purification on silica gel.The 1H
NMR spectrum [d 4.25 (d, J=7.5 Hz, H-1)] as well as
13C NMR spectrum [d 105.23 (C-1)] confirmed the for-
mation of b-linkage at C-1 (Scheme 1).
The synthesis of scaffold 1 from intermediate 10,
required that we oxidize the C-6 hydroxyl group to the
corresponding carboxylic acid in the presence of the
The synthesis of the 2,3,6-substituted scaffold 3 began
with previously prepared intermediate 17 (Scheme 4);
however, unlike in the preparation of scaffold 2, here we
Scheme 1. Reagents and conditions: (a) Me3SiBr, BiBr3, CH2Cl2, 4 h; (b) Ag2O, CaSO4, CH2Cl2, MeOH, 1 h; (c) MeOH, MeONa, 4 h.

R. Jain et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2185–2189
2187
Scheme 2. Reagents and conditions: (a) MeO–C6H4–CH(OMe)2, p-TSA, DMF, 4 h; (b) C5H5N, Ac2O, 4 h; (c) NaBH3CN, DMF, TFA, 3 A
molecular sieves, 48 h; (d) Ag2O, CH3I, DMF, 16 h; (e) DDQ, CH2Cl2 saturated with water, 2 h; (f) Jones reagent, acetone, 1.5 h; (g) MeOH–
MeONa, 4 h.
Scheme 3. Reagents and conditions: (a) Ag2O, BrCH2COOMe, KI, DMF, 16 h; (b) NaBH3CN, DMF, TFA, 3 A molecular sieves, 48 h; (c) Ag2O,
CH3I, DMF, 16 h; (d) DDQ, CH2Cl2 saturated with water, 2 h; (e) Dioxane, 2N NaOH, 1 h.
Scheme 4. Reagents and conditions: (a) NaBH3CN, Me3SiCl, 3 A molecular sieves, CH3CN, 4 h; (b) 21, Ag2O, CH3I, DMF, 16 h; (c) DDQ, CH2Cl2
saturated with water, 2 h; (d) dioxane, 2N NaOH, 1 h.
needed to achieve acetal ring opening that would favor
p-methoxybenzyl protection on the C-4 hydroxyl.This
was accomplished by treating compound 17 with
Me3SiCl–NaBH3CN11 in acetonitrile.These conditions
provide the 6-O-p-methoxybenzyl derivative 18 and
4-O-p-methoxybenzyl derivative 21 in 1:2 ratio.The C-6
hydroxyl group was then methylated using Ag2O and
methyl iodide and the p-methoxybenzyl group was
removed under oxidative conditions.Hydrolysis of the
methyl ester gave scaffold 3 in 57% overall yield.
opening and attachment of the carboxymethyl group
gave 25.Removal of the C-6 hydroxyl protecting group
and ester hydrolysis gave scaffold 4 in 38% overall
yield.
Synthesis of galactose scaffolds 5 and 6 from the gluco-
pyranosyl intermediate 18 required inverting the stereo-
chemistry at C-4.This inversion was accomplished
through intermediacy of the C-4 triflate 27 which was
obtained from 18 on treatment with triflic anhydride in
dichloromethane (Scheme 6).Nucleophilic displace-
ment of the C-4 triflate with tetrabutyl-ammonium
benzoate in toluene gave the axial benzoate 28 in 86%
For the preparation of scaffold 4, we needed to differ-
entiate the C-2, C-4, and C-6 hydroxyl groups as we had
done in the preparation of scaffold 1.Therefore, we again
used intermediate 11 and employed the regioselective
p-methoxybenzylidene ring opening protocol (Scheme
5).Starting with 11, treatment with CH3I-NaH in DMF
gave the C-2 methyl ether 23 in 95% yield.Acetal ring
yield.Removal of the
p-methoxybenzyl protecting
group gave 29.We next wanted to methylate the C-6
hydroxyl group; however, simultaneous transfer of the
benzoate group from the C-4–C-6 hydroxyl group and
methylation of the C-4 hydroxyl occurred on exposure

2188
R. Jain et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2185–2189
Scheme 5. Reagents and conditions: (a) NaH, DMF, CH3I, 1 h; (b) NaBH3CN, DMF, TFA, 3 A molecular sieves, 48 h; (c) Ag2O, BrCH2COOMe,
KI, DMF, 4 h; (d) DDQ, CH2Cl2 saturated with water, 2 h; (e) Dioxane, 2N NaOH,1 h.
Scheme 6. Reagents and conditions: (a) Triflic anhydride, pyridine, CH2Cl2, 0 ꢀC, 2 h; (b) Tetrabutylammonium benzoate, toluene, 1 h; (c) DDQ,
CH2Cl2 saturated with water, 2 h; (d) Ag2O, CH3I, DMF, 16 h; (e) Dioxane, 2N NaOH, 1 h.
Scheme 7. Reagents and conditions: (a) Triflic anhydride, pyridine, CH2Cl2, 0 ꢀC, 2 h; NaNO2, DMF, 4 h; (b) 95% aq.TFA, CH 2Cl2, 2 h; (c)
Trimethyloxonium tetrafluoroborate, 2,6-di-tert-butyl-4-methyl-pyridine, CH2Cl2, 16 h; (d) Dioxane, 2 N NaOH, 1 h.
to Ag2O/CH3I in DMF providing compound 30 in 94%
yield.With compound 30 in hand, base hydrolysis gave
scaffold 5 in 93% yield (Scheme 6).
graphy.Selective methylation of 31 with trimethyloxonium
tetrafluoroborate-2,6-ditertbutyl-4-methylpyridine13 gave
the 6-O-methyl derivative 32.The hydrolysis of methyl
ester provided 6 in 87% yield.
To access scaffold 6, triflate 27 was treated with sodium
nitrite12 in DMF providing the desired C-4 axial
hydroxyl group (Scheme 7).Subsequent removal of the
C-6 p-methoxybenzyl group with TFA–CH2Cl2 provided
diol 31 in 40% yield after silica gel column chromato-
To demonstrate the use of a 3-azido scaffold in the
construction of
saccharides we chose scaffold 2.Using Rink amide
a
library of substituted mono-
resin14 as the solid support, a blocked amino acid,
Scheme 8. Reagents and conditions: (a) (i) 20% piperidine, DMF, 0.5 h; (ii) Fmoc histidine (trt), HATU, DIEPA, DMF, 16 h; (b) 20% piperidine,
DMF, 0.5 h; (c) 2, HATU, DIEPA, DMF, 16 h; (d) isocyanate, THF, Et3N, 6–16 h or isocyanate, DMF, CuCl; (e) Me3P, THF/EtOH/H2O(4/4/1);
(f) carboxylic acid, HATU, DIEPA, DMF, 16 h; (g) 20% TFA, CH2Cl2, 2% Et3SiH.

R. Jain et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2185–2189
2189
Figure 2.
Fmoc-histidine (Trt), was first coupled to the support
(Scheme 8).Removal of the Fmoc protecting group
was then followed by coupling of scaffold 2 to the free
amino group by amide bond formation to the C-2 car-
boxymethyl group of 2.The resin-bound scaffold was
then treated with a series of isocyanates in the presence
of CuCl in DMF to generate the corresponding C-6
carbamates.Reduction of the C-3 azido group with
trimethylphosphine generated the desired amines which
were then reacted with a series of carboxylic acids to
provide the corresponding amides.
unique ligands for a wide variety of biomolecular drug
targets.To exploit this potential, we have developed a
series of 3-azido glycopyranosides.
References and Notes
1. Hirschmann, R.; Nicolaou, K. C.; Pietranico, S.; Leahy,
E. M.; Salvino, J.; Arison, B.; Cichy, M. A.; Spoors, P. G.;
Shakespear, W. C.; Sprengeler, P. A.; Hamley, P.; Smith,
A. B., III; Reisine, T.; Raynor, K.; Maechler, L.; Donaldson,
C.; Vale, W.; Freidinger, R. M.; Cascieri, M. R.; Strader, C. D.
J. Am. Chem. Soc. 1993, 115, 12550.
2.Hirschmann, R; Yao, W;. Cascieri, M.A;. Strader, C.D;.
Maechler, L.; Cichy-Knight, M. A.; Hynes, J., Jr; van Rijn, R. D.;
Sprengeler, P. A.; Smith, A. B., III J. Med. Chem. 1996, 39, 2441.
3. Wunberg, T.; Kallus, C.; Opatz, T.; Henke, S.; Schmidt,
W.; Kunz, H. Angew. Chemie. Int. Ed. Engl. 1998, 37, 2503.
4. Sofia, M. J.; Hunter, R.; Chan, T. Y.; Vaughan, A.; Dulina,
R.; Wang, H.; Gange, D. J. Org. Chem. 1998, 63, 2802.
5.Sofia, M. J. Med. Chem. Res. 1998, 8, 362.
A library of 48 compounds was prepared using the
directed sorting mix-and-split synthesis method15
(Fig.2 ).
All 48 compounds were analyzed by LC/MS using eva-
porative light-scattering detection (ELSD).The yield of
each product was determined from the ELSD trace
using standard curves and product identity was deter-
mined by electrospray mass spectal analysis.The aver-
age yield for the six-step sequence was 55%.One of the
library products was further purified by preparative
HPLC and its structure was confirmed by NMR spec-
troscopy and LC/MS analysis to be the desired com-
pound 40.16
6. Holla, E. W.; Sinnwell, V.; Klaffke, W. Synlett 1992, 5, 413.
7. Montero, J.-L.; Winum, J.-Y.; Leydet, A.; Kamal, M.;
Pavia, A. A.; Roque, J.-P. Carbohydr. Res. 1997, 297, 175.
8.Johansson, R;.Samuelsson, B. J. Chem. Soc., Chem. Com-
mun. 1984, 201.
9. Oikawa, Y.; Yshioka, T.; Yonemitsu, O. Tetrahedron Lett.
1982, 23, 885.
10. Allanson, N. M.; Liu, D.; Chi, F.; Jain, R. K.; Chen, A.;
Ghosh, M.; Hong, L.; Sofia, M. J. Tetrahedron Lett. 1998, 39,
1889.
11.Johansson, R;. Samuelsson, B.
Trans. 1 1994, 2317.
J. Chem. Soc., Perkin
12. El-Sokkary, R. I.; Silwanis, B. A.; Nashed, M. A. Carbo-
hydr. Res. 1990, 203, 319.
13.Fugedi, P;. Kovac, J.XV International Carbohydrate
Symposium, Aug.12–17, 1990, p 77.
14.Purchased from Novabiochem.
15.Radiofrequency tagging system.Nicolaou, K.C;. Xiao,
X. Y.; Parandoosh, Z.; Senyei, A.; Nova, M. P. Angew. Chem.
Int. Ed. Engl. 1995, 34, 2289.
16.NMR diagnostic signals at d 8.04 (d, 1H), 7.39 (s, 1H),
6.07 (s, 2H), 4.48 (d, 1H, J=7.5 Hz), 3.73 (s, 3H), 3.70 (s, 3H).
In conclusion, we believe that carbohydrate scaffolds
are biologically relevant molecular platforms that can
be used in combinatorial library strategies to identify