1,2,3-Triazolylalkylribitol derivatives as nucleoside hydrolase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.02.017

A range of novel 1,2,3-triazolylalkylribitol derivatives were synthesized and evaluated as nucleoside hydrolase inhibitors. The most active compound (11a) has low micromolar potency and is structurally diverse from previously reported nucleoside hydrolase

Full text of DOI:10.1016/j.bmcl.2007.02.017

Bioorganic & Medicinal Chemistry Letters 17 (2007) 2523–2526
1,2,3-Triazolylalkylribitol derivatives
as nucleoside hydrolase inhibitors
A. Goeminne,a M. McNaughton,a,* G. Bal,a G. Surpateanu,a P. Van der Veken,a
S. De Prol,b,c W. Versees, J. Steyaert,b,c S. Apers,d A. Haemersa and K. Augustynsa,
b,c
*
´
aDepartment of Medicinal Chemistry, University of Antwerp, Universiteitsplein 1, B-2610 Antwerp, Belgium
bDepartment of Ultrastructure, Vrije Universiteit Brussel, Pleinlaan 2, B-1050 Brussels, Belgium
cDepartment of Molecular Interactions,VIB, Pleinlaan 2, B-1050 Brussels, Belgium
dDepartment of Pharmacognosy, University of Antwerp, Universiteitsplein 1, B-2610 Antwerp, Belgium
Received 21 December 2006; revised 2 February 2007; accepted 7 February 2007
Available online 9 February 2007
Abstract—A range of novel 1,2,3-triazolylalkylribitol derivatives were synthesized and evaluated as nucleoside hydrolase inhibitors.
The most active compound (11a) has low micromolar potency and is structurally diverse from previously reported nucleoside hydro-
lase inhibitors, which, along with the simplicity of the chemistry involved in its synthesis, makes it a good lead for the further devel-
opment of novel nucleoside hydrolase inhibitors.
ꢀ 2007 Elsevier Ltd. All rights reserved.
African trypanosomiasis (sleeping sickness), American
trypanosomiasis (Chagas’ Disease) and leishmaniasis
are infections caused by parasitic protozoa from the Try-
panosomatidae family. These diseases continue to haunt
developing countries causing in excess of one hundred
thousand deaths annually.1 In order to combat these
infections, development of novel treatments is required.
Iminoribitol molecules, 1 and 2, originally developed by
Schramm et al.,5 have been reported with nanomolar
activity against the IAG-NH isolated from the parasite
T. vivax.6 Schramm et al. synthesized a range of structur-
ally less complex iminoribitol compounds and reported a
Ki = 3 lM against the IAG-NH from Trypanosoma brucei
brucei for the most active compound (3).7 However, imi-
noribitol-based IAG-NH inhibitors are mostly compro-
mised by their pronounced activity against nucleoside
phosphorylase, another ribonucleoside cleaving enzyme.
Moreover, there are a considerable number of synthetic
steps required to synthesize such iminoribitol com-
pounds. We were searching for IAG-NH inhibitors with
good activities, structurally distinct from iminoribitol-
based inhibitors and therefore synthetically easier to
access.
As part of our ongoing research into novel agents for the
treatment of trypanosomiasis we have been investigating
the nucleoside hydrolase enzyme as a potential target.
Parasitic protozoa are unable to synthesize purines de
novo and are reliant on the purine salvage pathway to
provide purinebases obtained from the nucleosides pres-
ent in the host.2 Nucleoside hydrolase (NH) is an essen-
tial enzyme in the purine salvage pathway. NH cleaves
the N-glycosidic bond of nucleosides sequestered from
the host to provide the purinebases which are necessary
for the survival of the parasite. Based on the substrate
specificity, four enzyme-types for NH are currently
known.3 Our target enzyme is isolated from the parasite
Trypanosoma vivax and is an IAG-NH, with a high spec-
ificity for inosine, adenosine and guanosine.4
Keywords: 1,2,3-Triazolylalkylribitol; 1,3-Dipolar cycloaddition;
Nucleoside hydrolase inhibitors.
It was known from modelling studies within our group
that ribitol derivatives with an aromatic or heteroaro-
matic substituent in the correct position can fit in the ac-
tive site of the target enzyme (IAG-NH from T. vivax,
*
Corresponding authors. Tel.: +3238202735/17; fax: +3238202739;
e-mail addresses: michael.mcnaughton@ua.ac.be; koen.augustyns@
ua.ac.be
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.02.017

2524
A. Goeminne et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2523–2526
Table 1. Inhibition of TvNH by the target compounds
TvNH) in an orientation that enables the aromatic sub-
stituent to participate in aromatic stacking interactions
with two tryptophan residues in the active site of TvNH.
We considered the triazolyl group as a good candidate
for these stacking interactions and in a search for bio-
logically active non-purine C-nucleosides, we synthe-
sized a series of 1,2,3-triazolylalkylribitol derivatives.
A 1,3-dipolar cycloaddition reaction was used with an
azidoalkylribitol as starting material.8 This would allow
for the possible creation of a wide range of ribitol deriv-
atives through the addition of appropriate alkynes to the
azidoalkylribitol. Herein we report our preliminary
findings.
Compound
Inhibition Ki (lM)
9a
9b
2.1 · 101 0.6 · 101
6.8 · 102 1.1 · 102
6.3 · 101 0.8 · 101
2.3 0.2
3.2 · 101 0.3 101
2.3 · 101 0.3 · 101
5.6 · 103 1.8 · 103
8.7 · 101 1.2 · 101
5.2 · 102 1.0 · 102
10
11a
11b
12
13
17a
17b
reaction times can be reduced with the addition of cop-
per (I) salts to the reaction conditions.11a
The synthesis of the ribitol derivatives started with the
azide-alkyl-ribitol 4 (Scheme 1), which can be readily
synthesized from D-ribose.9 Initially, phenylacetylene
was chosen as the first coupling partner for the azide.
Treatment of 4 with phenylacetylene in toluene at
80 ꢁC yielded the desired 1,2,3-triazole as its two possi-
ble regioisomers (5a:5b). At this stage the two isomers
could be easily separated by column chromatography.
Upon global deprotection of each isomer with trifluoro-
acetic acid, the desired ribitol compounds 9a and 9b
were obtained. The assignment of regiochemistry of 9a
and 9b was based on the literature where a difference
in chemical shift of approximately 0.5 ppm was ob-
served for the triazole-hydrogens of the different
isomers.10
Reaction of the selected alkynes with the azide 4 in the
presence of CuCl furnished the compounds 6–8 (Scheme
1). Global deprotection with trifluoroacetic acid fur-
nished the desired ribitol compounds 10–12. Synthetical-
ly, the 1,4-isomer was the sole isomer isolated in the case
of the terminal alkyne (12) as expected. The regiochem-
istry of 11a–b and 12 was assigned using 2D NMR tech-
niques (NOESY and HMBC experiments).
1,2,3-Triazole derivatives with only one carbon atom be-
tween the ribose and 1,2,3-triazole moieties were also
synthesized to observe how this would influence IAG-
NH inhibitor activities. The synthesis started with the
synthetically available amine 1412 which underwent a
diazo transfer reaction13 to furnish the required azide
(Scheme 2). Reaction with 1-phenyl-1-propyne yielded
the desired 1,2,3-triazole as the two possible isomers
which were deprotected with 7 N ammonia in methanol
to furnish 17a and 17b. Assignment of the relative regio-
chemistry was done by comparison of the 1D NMR
spectra of 17a–b with those of 11a–b.
At this stage it was decided to test these two 1,2,3-tria-
zoles for their activity as nucleoside hydrolase inhibitors
in order to see if there is a difference in activity between
the two isomers. It was revealed through the biochemi-
cal testing that there is a significant difference, with iso-
mer 9a being 30 times more active than 9b as an
inhibitor of IAG-NH (Table 1). A preliminary selection
of commercially available alkynes was subsequently
made in order to see if the activity could be improved.
Our target compounds were tested as inhibitors of
TvNH. A survey of the biochemical results as reported
in Table 1 reveals that the 1,2,3-triazolylalkyl group
on ribose considerably enhances inhibition of the nucle-
oside hydrolase compared to its free azide equivalent 13.
It is widely known that the addition of CuCl to the reac-
tion increases the yield of the desired 1,4-isomer in the
case of terminal alkynes.11 There are also reports that
It also appears from the biochemical results that the 4-
phenyl-triazole is the more active regioisomer and that
the addition of a methyl group at position 5 improves
potency. The most active compound in this series,
Scheme 2. Reagents and conditions: (a) triflic azide, NaHCO3,
CuSO4Æ5H2O, H2O/toluene/MeOH; (b) alkyne, toluene, CuCl,
130 ꢁC; (c) 7 N NH3 in MeOH.
Scheme 1. Reagents and conditions: (a) i—alkyne, toluene, 80 ꢁC, 24 h
or ii—alkyne, toluene, CuCl, 110–130 ꢁC; (b) TFA/H2O 1:1.

A. Goeminne et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2523–2526
2525
5-methyl-4-phenyl-1,2,3-triazolyliminoribitol derivative
11a, shows a Ki value of 2.26 lM. Shortening the alkyl
linker to one carbon, as in target compounds 17a–b,
considerably decreases potency.
whereas the 4-phenyl isomers were able to fit within the
active site. The compound with the best fit in the active site
is the most potent inhibitor of this series, namely 11a, and
docking of this inhibitor is illustrated in Figures 1b and 2.
The main interactions with the enzyme observed for our
inhibitors are illustrated in Figure 1b and they include
the interactions between the hydroxyl groups of the ribose
moiety, similar to what was observed for Immucillin-H.
In an attempt to find an explanation as to why the 4-
phenyl isomer is more active than the 5-phenyl isomer
we performed a molecular modelling study. In this
study, inhibitors 9a–b and 11a–b were docked in the ac-
tive site of the target enzyme, which was selected from
the pdb structure of TvNH co-crystallized with Immucil-
lin-H (2) (pdb code 2FF2).6 The structure of TvNH with
Immucillin-H is illustrated in Figure 1a.
Immucillin-H interacts with different active site residues
of the enzyme through the hydroxyl groups and the ring-
N of its iminoribitol moiety (Fig. 1a). The 9-deazahypo-
xanthine ring of Immucillin-H is orientated almost par-
allel between two Trp-residues (Trp83 and Trp260) of
the active site, allowing for aromatic stacking interac-
tions. The 9-deazahypoxanthine ring is also involved
in several hydrogen bonds: a first hydrogen bond is
formed between N7 and a water molecule in the active
site, a second hydrogen bond is formed between the
O6 carbonyl and amino acid residue Arg252, and a third
hydrogen bond is formed between N3 and amino acid
residue Asp40. The pKa of this Asp40 was calculated
to be approximately 7.2–7.8, making it mostly protonat-
ed in our experimental conditions (pH 7.0) and thus
allowing a hydrogen bond with N3.14
Figure 2. Binding of inhibitor 11a in the active site of the target
enzyme. The triazole moiety of the inhibitor is orientated parallel to
Trp83, enabling aromatic stacking interaction. Carbons are grey,
oxygens are red and nitrogens are blue. The colours of the enzyme
surface are according to the elements. The image was obtained with
PyMOL.15
An automatic docking protocol was used to position the
inhibitors into the active site. During the docking experi-
ments with ourinhibitors, itwas observed that the 5-phen-
yl isomers did not fit in the active site of the target enzyme,
Figure 1. (a) The active site of TvNH co-crystallized with Immucillin-H (2) (pdb code 2FF2). (b) Inhibitor 11a in the active site of the target enzyme.
For both images, the active site residues are depicted in grey (carbons), blue (nitrogens) and red (oxygens). The carbons of the inhibitors are depicted
in green, the colour code for nitrogens and oxygens is the same as for the active site residues. Water molecules are depicted as red spheres, the Ca2+
15
˚
ion is depicted as a magenta sphere. Distances are in A. The images were made with PyMOL.

2526
A. Goeminne et al. / Bioorg. Med. Chem. Lett. 17 (2007) 2523–2526
From Figure 2 it is clear that the aromatic substituent
(1,2,3-triazole moiety) is orientated parallel with one of
the two Trp-residues from the enzyme, which suggests
the possibility for aromatic stacking interactions. In the
orientation of compound 11a, as obtained during the
docking experiments and illustrated in Figure 2, the tria-
zole-nitrogens are pointing towards the protonated
Asp40-residue of the active site. This enables an interac-
tion between N2 from the triazole with Asp40.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.bmcl.2007.02.017.
References and notes
1. www.who.int/tdr.
2. El Kouni, M. H. Pharmacol. Ther. 2003, 99, 283.
´
3. Versees, W.; Steyaert, J. Curr. Opin. Struct. Biol. 2003, 13,
731.
Utilizing the commonly used 1,3-dipolar cycloaddition
reaction we have synthesized a range of 1,2,3-tria-
zolylalkylribitol derivatives with good to moderate
activities as inhibitors of IAG-NH. We have confirmed
that the 4-phenyl isomer is the more biochemically ac-
tive of the two isomers. The most active compound,
11a with Ki = 2.3 lM, is among the most active ribitol
inhibitors for IAG-NH reported and is synthetically eas-
ier to access than the previously reported iminoribitol
inhibitors. Compound 11a therefore provides us with a
good lead for the further development of non-iminorib-
itol IAG-NH inhibitors.
´
´
4. Versees, W.; Decanniere, K.; Pelle, R.; Depoorter, J.;
Brosens, E.; Parkin, D. W.; Steyaert, J. J. Mol. Biol. 2001,
307, 1363.
5. Miles, R. W.; Tyler, P. C.; Evans, G. B.; Furneaux, R. H.;
Schramm, V. L. Biochemistry 1999, 38, 13147.
´
6. Versees, W.; Barlow, J.; Steyeart, J. J. Mol. Biol. 2006,
359, 331.
7. Furneaux, R. H.; Schramm, V. L.; Tyler, P. C. Bioorg.
Med. Chem. 1999, 7, 2599; Parkin, D. W.; Limberg, G.;
Tyler, P. C.; Furneaux, R. H.; Chen, X.-Y.; Schramm, V.
L. Biochemistry 1997, 36, 3528.
8. Kolb, H. C.; Sharpless, K. B. Drug Discovery Today 2003,
8, 1128.
9. See Supplementary Data.
10. Blass, B. E.; Coburn, K. R.; Faulkner, A. L.; Hunn, C.
L.; Natchus, M. G.; Sharon Parker, M.; Portlock, D.
E.; Tullis, J. S.; Wood, R. Tetrahedron Lett. 2002, 43,
4059.
11. (a) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.;
Sharpless, K. B. Angew. Chem. Int. Ed. 2002, 41, 2596;
(b) Tornøe, C. W.; Christensen, C.; Meldal, M. J. Org.
Chem. 2002, 67, 3057.
Acknowledgments
This work was funded by a research project from the
Institute for the Promotion of Innovation through Sci-
ence and Technology in Flanders (IWT-Vlaanderen).
A.G. has a Ph.D. grant from the same institute. W.V.
and P. VDV. have a postdoctoral grant from the Fund
for Scientific Research-Flanders (FWO-Vlaanderen).
´
12. de las Heras, F. G.; Fernandez-Resa, P. J. Chem. Soc.,
Perkin Trans. 1 1982, 903; Sauer, D. R.; Schneller, S. W.
Synthesis 1991, 747.
13. Alper, P. B.; Hung, S.-C.; Wong, C.-H. Tetrahedron Lett.
1996, 37, 6029.
Supplementary data
14. Mazumder-Shivakumar, D.; Bruice, T. C. Biochemistry
2005, 44, 7805.
15. PyMOL Molecular Graphics System, DeLano Scientific,
San Carlos, CA.
Detailed synthetic procedures and analytical data for the
target compounds and biochemical protocol for Ki-de-
termination are given.